All Downloads are FREE. Search and download functionalities are using the official Maven repository.

io.reactivex.Flowable Maven / Gradle / Ivy

There is a newer version: 2.2.21
Show newest version
/**
 * Copyright (c) 2016-present, RxJava Contributors.
 *
 * Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in
 * compliance with the License. You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software distributed under the License is
 * distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See
 * the License for the specific language governing permissions and limitations under the License.
 */
package io.reactivex;

import java.util.*;
import java.util.concurrent.*;

import org.reactivestreams.*;

import io.reactivex.annotations.*;
import io.reactivex.disposables.Disposable;
import io.reactivex.exceptions.Exceptions;
import io.reactivex.flowables.*;
import io.reactivex.functions.*;
import io.reactivex.internal.functions.*;
import io.reactivex.internal.fuseable.*;
import io.reactivex.internal.operators.flowable.*;
import io.reactivex.internal.operators.mixed.*;
import io.reactivex.internal.operators.observable.*;
import io.reactivex.internal.schedulers.ImmediateThinScheduler;
import io.reactivex.internal.subscribers.*;
import io.reactivex.internal.util.*;
import io.reactivex.parallel.ParallelFlowable;
import io.reactivex.plugins.RxJavaPlugins;
import io.reactivex.schedulers.*;
import io.reactivex.subscribers.*;

/**
 * The Flowable class that implements the Reactive Streams
 * Pattern and offers factory methods, intermediate operators and the ability to consume reactive dataflows.
 * 

* Reactive Streams operates with {@link Publisher}s which {@code Flowable} extends. Many operators * therefore accept general {@code Publisher}s directly and allow direct interoperation with other * Reactive Streams implementations. *

* The Flowable hosts the default buffer size of 128 elements for operators, accessible via {@link #bufferSize()}, * that can be overridden globally via the system parameter {@code rx2.buffer-size}. Most operators, however, have * overloads that allow setting their internal buffer size explicitly. *

* The documentation for this class makes use of marble diagrams. The following legend explains these diagrams: *

* *

* The {@code Flowable} follows the protocol *


 *      onSubscribe onNext* (onError | onComplete)?
 * 
* where the stream can be disposed through the {@link Subscription} instance provided to consumers through * {@link Subscriber#onSubscribe(Subscription)}. * Unlike the {@code Observable.subscribe()} of version 1.x, {@link #subscribe(Subscriber)} does not allow external cancellation * of a subscription and the {@link Subscriber} instance is expected to expose such capability if needed. *

* Flowables support backpressure and require {@link Subscriber}s to signal demand via {@link Subscription#request(long)}. *

* Example: *


 * Disposable d = Flowable.just("Hello world!")
 *     .delay(1, TimeUnit.SECONDS)
 *     .subscribeWith(new DisposableSubscriber<String>() {
 *         @Override public void onStart() {
 *             System.out.println("Start!");
 *             request(1);
 *         }
 *         @Override public void onNext(String t) {
 *             System.out.println(t);
 *             request(1);
 *         }
 *         @Override public void onError(Throwable t) {
 *             t.printStackTrace();
 *         }
 *         @Override public void onComplete() {
 *             System.out.println("Done!");
 *         }
 *     });
 *
 * Thread.sleep(500);
 * // the sequence can now be cancelled via dispose()
 * d.dispose();
 * 
*

* The Reactive Streams specification is relatively strict when defining interactions between {@code Publisher}s and {@code Subscriber}s, so much so * that there is a significant performance penalty due certain timing requirements and the need to prepare for invalid * request amounts via {@link Subscription#request(long)}. * Therefore, RxJava has introduced the {@link FlowableSubscriber} interface that indicates the consumer can be driven with relaxed rules. * All RxJava operators are implemented with these relaxed rules in mind. * If the subscribing {@code Subscriber} does not implement this interface, for example, due to it being from another Reactive Streams compliant * library, the Flowable will automatically apply a compliance wrapper around it. *

* {@code Flowable} is an abstract class, but it is not advised to implement sources and custom operators by extending the class directly due * to the large amounts of Reactive Streams * rules to be followed to the letter. See the wiki for * some guidance if such custom implementations are necessary. *

* The recommended way of creating custom {@code Flowable}s is by using the {@link #create(FlowableOnSubscribe, BackpressureStrategy)} factory method: *


 * Flowable<String> source = Flowable.create(new FlowableOnSubscribe<String>() {
 *     @Override
 *     public void subscribe(FlowableEmitter<String> emitter) throws Exception {
 *
 *         // signal an item
 *         emitter.onNext("Hello");
 *
 *         // could be some blocking operation
 *         Thread.sleep(1000);
 *
 *         // the consumer might have cancelled the flow
 *         if (emitter.isCancelled() {
 *             return;
 *         }
 *
 *         emitter.onNext("World");
 *
 *         Thread.sleep(1000);
 *
 *         // the end-of-sequence has to be signaled, otherwise the
 *         // consumers may never finish
 *         emitter.onComplete();
 *     }
 * }, BackpressureStrategy.BUFFER);
 *
 * System.out.println("Subscribe!");
 * 
 * source.subscribe(System.out::println);
 * 
 * System.out.println("Done!");
 * 
*

* RxJava reactive sources, such as {@code Flowable}, are generally synchronous and sequential in nature. In the ReactiveX design, the location (thread) * where operators run is orthogonal to when the operators can work with data. This means that asynchrony and parallelism * has to be explicitly expressed via operators such as {@link #subscribeOn(Scheduler)}, {@link #observeOn(Scheduler)} and {@link #parallel()}. In general, * operators featuring a {@link Scheduler} parameter are introducing this type of asynchrony into the flow. *

* For more information see the ReactiveX * documentation. * * @param * the type of the items emitted by the Flowable * @see Observable * @see ParallelFlowable * @see io.reactivex.subscribers.DisposableSubscriber */ public abstract class Flowable implements Publisher { /** The default buffer size. */ static final int BUFFER_SIZE; static { BUFFER_SIZE = Math.max(1, Integer.getInteger("rx2.buffer-size", 128)); } /** * Mirrors the one Publisher in an Iterable of several Publishers that first either emits an item or sends * a termination notification. *

* *

*
Backpressure:
*
The operator itself doesn't interfere with backpressure which is determined by the winning * {@code Publisher}'s backpressure behavior.
*
Scheduler:
*
{@code amb} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element type * @param sources * an Iterable of Publishers sources competing to react first. A subscription to each Publisher will * occur in the same order as in this Iterable. * @return a Flowable that emits the same sequence as whichever of the source Publishers first * emitted an item or sent a termination notification * @see ReactiveX operators documentation: Amb */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable amb(Iterable> sources) { ObjectHelper.requireNonNull(sources, "sources is null"); return RxJavaPlugins.onAssembly(new FlowableAmb(null, sources)); } /** * Mirrors the one Publisher in an array of several Publishers that first either emits an item or sends * a termination notification. *

* *

*
Backpressure:
*
The operator itself doesn't interfere with backpressure which is determined by the winning * {@code Publisher}'s backpressure behavior.
*
Scheduler:
*
{@code ambArray} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element type * @param sources * an array of Publisher sources competing to react first. A subscription to each Publisher will * occur in the same order as in this Iterable. * @return a Flowable that emits the same sequence as whichever of the source Publishers first * emitted an item or sent a termination notification * @see ReactiveX operators documentation: Amb */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable ambArray(Publisher... sources) { ObjectHelper.requireNonNull(sources, "sources is null"); int len = sources.length; if (len == 0) { return empty(); } else if (len == 1) { return fromPublisher(sources[0]); } return RxJavaPlugins.onAssembly(new FlowableAmb(sources, null)); } /** * Returns the default internal buffer size used by most async operators. *

The value can be overridden via system parameter {@code rx2.buffer-size} * before the Flowable class is loaded. * @return the default internal buffer size. */ public static int bufferSize() { return BUFFER_SIZE; } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting * any items and without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatest(Publisher[] sources, Function combiner) { return combineLatest(sources, combiner, bufferSize()); } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If there are no source Publishers provided, the resulting sequence completes immediately without emitting * any items and without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatest(Function combiner, Publisher... sources) { return combineLatest(sources, combiner, bufferSize()); } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting * any items and without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @param bufferSize * the internal buffer size and prefetch amount applied to every source Flowable * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatest(Publisher[] sources, Function combiner, int bufferSize) { ObjectHelper.requireNonNull(sources, "sources is null"); if (sources.length == 0) { return empty(); } ObjectHelper.requireNonNull(combiner, "combiner is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableCombineLatest(sources, combiner, bufferSize, false)); } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting * any items and without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatest(Iterable> sources, Function combiner) { return combineLatest(sources, combiner, bufferSize()); } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting any items and * without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @param bufferSize * the internal buffer size and prefetch amount applied to every source Flowable * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatest(Iterable> sources, Function combiner, int bufferSize) { ObjectHelper.requireNonNull(sources, "sources is null"); ObjectHelper.requireNonNull(combiner, "combiner is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableCombineLatest(sources, combiner, bufferSize, false)); } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting * any items and without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatestDelayError(Publisher[] sources, Function combiner) { return combineLatestDelayError(sources, combiner, bufferSize()); } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function and delays any error from the sources until * all source Publishers terminate. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If there are no source Publishers provided, the resulting sequence completes immediately without emitting * any items and without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatestDelayError(Function combiner, Publisher... sources) { return combineLatestDelayError(sources, combiner, bufferSize()); } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publisher, where this * aggregation is defined by a specified function and delays any error from the sources until * all source Publishers terminate. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If there are no source Publishers provided, the resulting sequence completes immediately without emitting * any items and without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @param bufferSize * the internal buffer size and prefetch amount applied to every source Publisher * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatestDelayError(Function combiner, int bufferSize, Publisher... sources) { return combineLatestDelayError(sources, combiner, bufferSize); } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function and delays any error from the sources until * all source Publishers terminate. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting * any items and without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @param bufferSize * the internal buffer size and prefetch amount applied to every source Flowable * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatestDelayError(Publisher[] sources, Function combiner, int bufferSize) { ObjectHelper.requireNonNull(sources, "sources is null"); ObjectHelper.requireNonNull(combiner, "combiner is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); if (sources.length == 0) { return empty(); } return RxJavaPlugins.onAssembly(new FlowableCombineLatest(sources, combiner, bufferSize, true)); } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function and delays any error from the sources until * all source Publishers terminate. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting * any items and without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatestDelayError(Iterable> sources, Function combiner) { return combineLatestDelayError(sources, combiner, bufferSize()); } /** * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of * the source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function and delays any error from the sources until * all source Publishers terminate. *

* Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a * {@code Function} passed to the method would trigger a {@code ClassCastException}. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting * any items and without any calls to the combiner function. * *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the common base type of source values * @param * the result type * @param sources * the collection of source Publishers * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @param bufferSize * the internal buffer size and prefetch amount applied to every source Flowable * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) public static Flowable combineLatestDelayError(Iterable> sources, Function combiner, int bufferSize) { ObjectHelper.requireNonNull(sources, "sources is null"); ObjectHelper.requireNonNull(combiner, "combiner is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableCombineLatest(sources, combiner, bufferSize, true)); } /** * Combines two source Publishers by emitting an item that aggregates the latest values of each of the * source Publishers each time an item is received from either of the source Publishers, where this * aggregation is defined by a specified function. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the first source * @param the element type of the second source * @param the combined output type * @param source1 * the first source Publisher * @param source2 * the second source Publisher * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SuppressWarnings("unchecked") @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable combineLatest( Publisher source1, Publisher source2, BiFunction combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); Function f = Functions.toFunction(combiner); return combineLatest(f, source1, source2); } /** * Combines three source Publishers by emitting an item that aggregates the latest values of each of the * source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the first source * @param the element type of the second source * @param the element type of the third source * @param the combined output type * @param source1 * the first source Publisher * @param source2 * the second source Publisher * @param source3 * the third source Publisher * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable combineLatest( Publisher source1, Publisher source2, Publisher source3, Function3 combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); return combineLatest(Functions.toFunction(combiner), source1, source2, source3); } /** * Combines four source Publishers by emitting an item that aggregates the latest values of each of the * source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the first source * @param the element type of the second source * @param the element type of the third source * @param the element type of the fourth source * @param the combined output type * @param source1 * the first source Publisher * @param source2 * the second source Publisher * @param source3 * the third source Publisher * @param source4 * the fourth source Publisher * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable combineLatest( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Function4 combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4); } /** * Combines five source Publishers by emitting an item that aggregates the latest values of each of the * source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the first source * @param the element type of the second source * @param the element type of the third source * @param the element type of the fourth source * @param the element type of the fifth source * @param the combined output type * @param source1 * the first source Publisher * @param source2 * the second source Publisher * @param source3 * the third source Publisher * @param source4 * the fourth source Publisher * @param source5 * the fifth source Publisher * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable combineLatest( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Publisher source5, Function5 combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); ObjectHelper.requireNonNull(source5, "source5 is null"); return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4, source5); } /** * Combines six source Publishers by emitting an item that aggregates the latest values of each of the * source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the first source * @param the element type of the second source * @param the element type of the third source * @param the element type of the fourth source * @param the element type of the fifth source * @param the element type of the sixth source * @param the combined output type * @param source1 * the first source Publisher * @param source2 * the second source Publisher * @param source3 * the third source Publisher * @param source4 * the fourth source Publisher * @param source5 * the fifth source Publisher * @param source6 * the sixth source Publisher * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable combineLatest( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Publisher source5, Publisher source6, Function6 combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); ObjectHelper.requireNonNull(source5, "source5 is null"); ObjectHelper.requireNonNull(source6, "source6 is null"); return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4, source5, source6); } /** * Combines seven source Publishers by emitting an item that aggregates the latest values of each of the * source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the first source * @param the element type of the second source * @param the element type of the third source * @param the element type of the fourth source * @param the element type of the fifth source * @param the element type of the sixth source * @param the element type of the seventh source * @param the combined output type * @param source1 * the first source Publisher * @param source2 * the second source Publisher * @param source3 * the third source Publisher * @param source4 * the fourth source Publisher * @param source5 * the fifth source Publisher * @param source6 * the sixth source Publisher * @param source7 * the seventh source Publisher * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable combineLatest( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Publisher source5, Publisher source6, Publisher source7, Function7 combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); ObjectHelper.requireNonNull(source5, "source5 is null"); ObjectHelper.requireNonNull(source6, "source6 is null"); ObjectHelper.requireNonNull(source7, "source7 is null"); return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4, source5, source6, source7); } /** * Combines eight source Publishers by emitting an item that aggregates the latest values of each of the * source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the first source * @param the element type of the second source * @param the element type of the third source * @param the element type of the fourth source * @param the element type of the fifth source * @param the element type of the sixth source * @param the element type of the seventh source * @param the element type of the eighth source * @param the combined output type * @param source1 * the first source Publisher * @param source2 * the second source Publisher * @param source3 * the third source Publisher * @param source4 * the fourth source Publisher * @param source5 * the fifth source Publisher * @param source6 * the sixth source Publisher * @param source7 * the seventh source Publisher * @param source8 * the eighth source Publisher * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable combineLatest( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Publisher source5, Publisher source6, Publisher source7, Publisher source8, Function8 combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); ObjectHelper.requireNonNull(source5, "source5 is null"); ObjectHelper.requireNonNull(source6, "source6 is null"); ObjectHelper.requireNonNull(source7, "source7 is null"); ObjectHelper.requireNonNull(source8, "source8 is null"); return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4, source5, source6, source7, source8); } /** * Combines nine source Publishers by emitting an item that aggregates the latest values of each of the * source Publishers each time an item is received from any of the source Publishers, where this * aggregation is defined by a specified function. *

* If any of the sources never produces an item but only terminates (normally or with an error), the * resulting sequence terminates immediately (normally or with all the errors accumulated until that point). * If that input source is also synchronous, other sources after it will not be subscribed to. *

* *

*
Backpressure:
*
The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s * are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal * {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code combineLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the first source * @param the element type of the second source * @param the element type of the third source * @param the element type of the fourth source * @param the element type of the fifth source * @param the element type of the sixth source * @param the element type of the seventh source * @param the element type of the eighth source * @param the element type of the ninth source * @param the combined output type * @param source1 * the first source Publisher * @param source2 * the second source Publisher * @param source3 * the third source Publisher * @param source4 * the fourth source Publisher * @param source5 * the fifth source Publisher * @param source6 * the sixth source Publisher * @param source7 * the seventh source Publisher * @param source8 * the eighth source Publisher * @param source9 * the ninth source Publisher * @param combiner * the aggregation function used to combine the items emitted by the source Publishers * @return a Flowable that emits items that are the result of combining the items emitted by the source * Publishers by means of the given aggregation function * @see ReactiveX operators documentation: CombineLatest */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable combineLatest( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Publisher source5, Publisher source6, Publisher source7, Publisher source8, Publisher source9, Function9 combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); ObjectHelper.requireNonNull(source5, "source5 is null"); ObjectHelper.requireNonNull(source6, "source6 is null"); ObjectHelper.requireNonNull(source7, "source7 is null"); ObjectHelper.requireNonNull(source8, "source8 is null"); ObjectHelper.requireNonNull(source9, "source9 is null"); return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4, source5, source6, source7, source8, source9); } /** * Concatenates elements of each Publisher provided via an Iterable sequence into a single sequence * of elements without interleaving them. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code concat} does not operate by default on a particular {@link Scheduler}.
*
* @param the common value type of the sources * @param sources the Iterable sequence of Publishers * @return the new Flowable instance */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concat(Iterable> sources) { ObjectHelper.requireNonNull(sources, "sources is null"); // unlike general sources, fromIterable can only throw on a boundary because it is consumed only there return fromIterable(sources).concatMapDelayError((Function)Functions.identity(), 2, false); } /** * Returns a Flowable that emits the items emitted by each of the Publishers emitted by the source * Publisher, one after the other, without interleaving them. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. Both the outer and inner {@code Publisher} * sources are expected to honor backpressure as well. If the outer violates this, a * {@code MissingBackpressureException} is signaled. If any of the inner {@code Publisher}s violates * this, it may throw an {@code IllegalStateException} when an inner {@code Publisher} completes.
*
Scheduler:
*
{@code concat} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources * a Publisher that emits Publishers * @return a Flowable that emits items all of the items emitted by the Publishers emitted by * {@code Publishers}, one after the other, without interleaving them * @see ReactiveX operators documentation: Concat */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concat(Publisher> sources) { return concat(sources, bufferSize()); } /** * Returns a Flowable that emits the items emitted by each of the Publishers emitted by the source * Publisher, one after the other, without interleaving them. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. Both the outer and inner {@code Publisher} * sources are expected to honor backpressure as well. If the outer violates this, a * {@code MissingBackpressureException} is signaled. If any of the inner {@code Publisher}s violates * this, it may throw an {@code IllegalStateException} when an inner {@code Publisher} completes.
*
Scheduler:
*
{@code concat} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources * a Publisher that emits Publishers * @param prefetch * the number of Publishers to prefetch from the sources sequence. * @return a Flowable that emits items all of the items emitted by the Publishers emitted by * {@code Publishers}, one after the other, without interleaving them * @see ReactiveX operators documentation: Concat */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concat(Publisher> sources, int prefetch) { return fromPublisher(sources).concatMap((Function)Functions.identity(), prefetch); } /** * Returns a Flowable that emits the items emitted by two Publishers, one after the other, without * interleaving them. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code concat} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param source1 * a Publisher to be concatenated * @param source2 * a Publisher to be concatenated * @return a Flowable that emits items emitted by the two source Publishers, one after the other, * without interleaving them * @see ReactiveX operators documentation: Concat */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concat(Publisher source1, Publisher source2) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); return concatArray(source1, source2); } /** * Returns a Flowable that emits the items emitted by three Publishers, one after the other, without * interleaving them. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code concat} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param source1 * a Publisher to be concatenated * @param source2 * a Publisher to be concatenated * @param source3 * a Publisher to be concatenated * @return a Flowable that emits items emitted by the three source Publishers, one after the other, * without interleaving them * @see ReactiveX operators documentation: Concat */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concat( Publisher source1, Publisher source2, Publisher source3) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); return concatArray(source1, source2, source3); } /** * Returns a Flowable that emits the items emitted by four Publishers, one after the other, without * interleaving them. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code concat} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param source1 * a Publisher to be concatenated * @param source2 * a Publisher to be concatenated * @param source3 * a Publisher to be concatenated * @param source4 * a Publisher to be concatenated * @return a Flowable that emits items emitted by the four source Publishers, one after the other, * without interleaving them * @see ReactiveX operators documentation: Concat */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concat( Publisher source1, Publisher source2, Publisher source3, Publisher source4) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); return concatArray(source1, source2, source3, source4); } /** * Concatenates a variable number of Publisher sources. *

* Note: named this way because of overload conflict with concat(Publisher<Publisher>). *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code concatArray} does not operate by default on a particular {@link Scheduler}.
*
* @param sources the array of sources * @param the common base value type * @return the new Publisher instance * @throws NullPointerException if sources is null */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concatArray(Publisher... sources) { if (sources.length == 0) { return empty(); } else if (sources.length == 1) { return fromPublisher(sources[0]); } return RxJavaPlugins.onAssembly(new FlowableConcatArray(sources, false)); } /** * Concatenates a variable number of Publisher sources and delays errors from any of them * till all terminate. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code concatArrayDelayError} does not operate by default on a particular {@link Scheduler}.
*
* @param sources the array of sources * @param the common base value type * @return the new Flowable instance * @throws NullPointerException if sources is null */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concatArrayDelayError(Publisher... sources) { if (sources.length == 0) { return empty(); } else if (sources.length == 1) { return fromPublisher(sources[0]); } return RxJavaPlugins.onAssembly(new FlowableConcatArray(sources, true)); } /** * Concatenates an array of Publishers eagerly into a single stream of values. *

* *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * source Publishers. The operator buffers the values emitted by these Publishers and then drains them * in order, each one after the previous one completes. *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, the operator will signal a * {@code MissingBackpressureException}.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param sources an array of Publishers that need to be eagerly concatenated * @return the new Publisher instance with the specified concatenation behavior * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concatArrayEager(Publisher... sources) { return concatArrayEager(bufferSize(), bufferSize(), sources); } /** * Concatenates an array of Publishers eagerly into a single stream of values. *

* *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * source Publishers. The operator buffers the values emitted by these Publishers and then drains them * in order, each one after the previous one completes. *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, the operator will signal a * {@code MissingBackpressureException}.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param sources an array of Publishers that need to be eagerly concatenated * @param maxConcurrency the maximum number of concurrent subscriptions at a time, Integer.MAX_VALUE * is interpreted as an indication to subscribe to all sources at once * @param prefetch the number of elements to prefetch from each Publisher source * @return the new Publisher instance with the specified concatenation behavior * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) @SuppressWarnings({ "rawtypes", "unchecked" }) public static Flowable concatArrayEager(int maxConcurrency, int prefetch, Publisher... sources) { ObjectHelper.requireNonNull(sources, "sources is null"); ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapEager(new FlowableFromArray(sources), Functions.identity(), maxConcurrency, prefetch, ErrorMode.IMMEDIATE)); } /** * Concatenates an array of {@link Publisher}s eagerly into a single stream of values * and delaying any errors until all sources terminate. *

* *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * source {@code Publisher}s. The operator buffers the values emitted by these {@code Publisher}s * and then drains them in order, each one after the previous one completes. *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, the operator will signal a * {@code MissingBackpressureException}.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param sources an array of {@code Publisher}s that need to be eagerly concatenated * @return the new Flowable instance with the specified concatenation behavior * @since 2.2.1 - experimental */ @CheckReturnValue @SchedulerSupport(SchedulerSupport.NONE) @BackpressureSupport(BackpressureKind.FULL) public static Flowable concatArrayEagerDelayError(Publisher... sources) { return concatArrayEagerDelayError(bufferSize(), bufferSize(), sources); } /** * Concatenates an array of {@link Publisher}s eagerly into a single stream of values * and delaying any errors until all sources terminate. *

* *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * source {@code Publisher}s. The operator buffers the values emitted by these {@code Publisher}s * and then drains them in order, each one after the previous one completes. *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, the operator will signal a * {@code MissingBackpressureException}.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param sources an array of {@code Publisher}s that need to be eagerly concatenated * @param maxConcurrency the maximum number of concurrent subscriptions at a time, Integer.MAX_VALUE * is interpreted as indication to subscribe to all sources at once * @param prefetch the number of elements to prefetch from each {@code Publisher} source * @return the new Flowable instance with the specified concatenation behavior * @since 2.2.1 - experimental */ @SuppressWarnings({ "rawtypes", "unchecked" }) @CheckReturnValue @SchedulerSupport(SchedulerSupport.NONE) @BackpressureSupport(BackpressureKind.FULL) public static Flowable concatArrayEagerDelayError(int maxConcurrency, int prefetch, Publisher... sources) { return fromArray(sources).concatMapEagerDelayError((Function)Functions.identity(), maxConcurrency, prefetch, true); } /** * Concatenates the Iterable sequence of Publishers into a single sequence by subscribing to each Publisher, * one after the other, one at a time and delays any errors till the all inner Publishers terminate. * *
*
Backpressure:
*
The operator honors backpressure from downstream. Both the outer and inner {@code Publisher} * sources are expected to honor backpressure as well. If the outer violates this, a * {@code MissingBackpressureException} is signaled. If any of the inner {@code Publisher}s violates * this, it may throw an {@code IllegalStateException} when an inner {@code Publisher} completes.
*
Scheduler:
*
{@code concatDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources the Iterable sequence of Publishers * @return the new Publisher with the concatenating behavior */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concatDelayError(Iterable> sources) { ObjectHelper.requireNonNull(sources, "sources is null"); return fromIterable(sources).concatMapDelayError((Function)Functions.identity()); } /** * Concatenates the Publisher sequence of Publishers into a single sequence by subscribing to each inner Publisher, * one after the other, one at a time and delays any errors till the all inner and the outer Publishers terminate. * *
*
Backpressure:
*
{@code concatDelayError} fully supports backpressure.
*
Scheduler:
*
{@code concatDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources the Publisher sequence of Publishers * @return the new Publisher with the concatenating behavior */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concatDelayError(Publisher> sources) { return concatDelayError(sources, bufferSize(), true); } /** * Concatenates the Publisher sequence of Publishers into a single sequence by subscribing to each inner Publisher, * one after the other, one at a time and delays any errors till the all inner and the outer Publishers terminate. * *
*
Backpressure:
*
{@code concatDelayError} fully supports backpressure.
*
Scheduler:
*
{@code concatDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources the Publisher sequence of Publishers * @param prefetch the number of elements to prefetch from the outer Publisher * @param tillTheEnd if true exceptions from the outer and all inner Publishers are delayed to the end * if false, exception from the outer Publisher is delayed till the current Publisher terminates * @return the new Publisher with the concatenating behavior */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concatDelayError(Publisher> sources, int prefetch, boolean tillTheEnd) { return fromPublisher(sources).concatMapDelayError((Function)Functions.identity(), prefetch, tillTheEnd); } /** * Concatenates a Publisher sequence of Publishers eagerly into a single stream of values. *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * emitted source Publishers as they are observed. The operator buffers the values emitted by these * Publishers and then drains them in order, each one after the previous one completes. *

*
Backpressure:
*
Backpressure is honored towards the downstream and both the outer and inner Publishers are * expected to support backpressure. Violating this assumption, the operator will * signal {@code MissingBackpressureException}.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param sources a sequence of Publishers that need to be eagerly concatenated * @return the new Publisher instance with the specified concatenation behavior * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concatEager(Publisher> sources) { return concatEager(sources, bufferSize(), bufferSize()); } /** * Concatenates a Publisher sequence of Publishers eagerly into a single stream of values. *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * emitted source Publishers as they are observed. The operator buffers the values emitted by these * Publishers and then drains them in order, each one after the previous one completes. *

*
Backpressure:
*
Backpressure is honored towards the downstream and both the outer and inner Publishers are * expected to support backpressure. Violating this assumption, the operator will * signal {@code MissingBackpressureException}.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param sources a sequence of Publishers that need to be eagerly concatenated * @param maxConcurrency the maximum number of concurrently running inner Publishers; Integer.MAX_VALUE * is interpreted as all inner Publishers can be active at the same time * @param prefetch the number of elements to prefetch from each inner Publisher source * @return the new Publisher instance with the specified concatenation behavior * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) @SuppressWarnings({ "rawtypes", "unchecked" }) public static Flowable concatEager(Publisher> sources, int maxConcurrency, int prefetch) { ObjectHelper.requireNonNull(sources, "sources is null"); ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapEagerPublisher(sources, Functions.identity(), maxConcurrency, prefetch, ErrorMode.IMMEDIATE)); } /** * Concatenates a sequence of Publishers eagerly into a single stream of values. *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * source Publishers. The operator buffers the values emitted by these Publishers and then drains them * in order, each one after the previous one completes. *

*
Backpressure:
*
Backpressure is honored towards the downstream and the inner Publishers are * expected to support backpressure. Violating this assumption, the operator will * signal {@code MissingBackpressureException}.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param sources a sequence of Publishers that need to be eagerly concatenated * @return the new Publisher instance with the specified concatenation behavior * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable concatEager(Iterable> sources) { return concatEager(sources, bufferSize(), bufferSize()); } /** * Concatenates a sequence of Publishers eagerly into a single stream of values. *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * source Publishers. The operator buffers the values emitted by these Publishers and then drains them * in order, each one after the previous one completes. *

*
Backpressure:
*
Backpressure is honored towards the downstream and both the outer and inner Publishers are * expected to support backpressure. Violating this assumption, the operator will * signal {@code MissingBackpressureException}.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param sources a sequence of Publishers that need to be eagerly concatenated * @param maxConcurrency the maximum number of concurrently running inner Publishers; Integer.MAX_VALUE * is interpreted as all inner Publishers can be active at the same time * @param prefetch the number of elements to prefetch from each inner Publisher source * @return the new Publisher instance with the specified concatenation behavior * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) @SuppressWarnings({ "rawtypes", "unchecked" }) public static Flowable concatEager(Iterable> sources, int maxConcurrency, int prefetch) { ObjectHelper.requireNonNull(sources, "sources is null"); ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapEager(new FlowableFromIterable(sources), Functions.identity(), maxConcurrency, prefetch, ErrorMode.IMMEDIATE)); } /** * Provides an API (via a cold Flowable) that bridges the reactive world with the callback-style, * generally non-backpressured world. *

* Example: *


     * Flowable.<Event>create(emitter -> {
     *     Callback listener = new Callback() {
     *         @Override
     *         public void onEvent(Event e) {
     *             emitter.onNext(e);
     *             if (e.isLast()) {
     *                 emitter.onComplete();
     *             }
     *         }
     *
     *         @Override
     *         public void onFailure(Exception e) {
     *             emitter.onError(e);
     *         }
     *     };
     *
     *     AutoCloseable c = api.someMethod(listener);
     *
     *     emitter.setCancellable(c::close);
     *
     * }, BackpressureStrategy.BUFFER);
     * 
*

* You should call the FlowableEmitter onNext, onError and onComplete methods in a serialized fashion. The * rest of its methods are thread-safe. *

*
Backpressure:
*
The backpressure behavior is determined by the {@code mode} parameter.
*
Scheduler:
*
{@code create} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type * @param source the emitter that is called when a Subscriber subscribes to the returned {@code Flowable} * @param mode the backpressure mode to apply if the downstream Subscriber doesn't request (fast) enough * @return the new Flowable instance * @see FlowableOnSubscribe * @see BackpressureStrategy * @see Cancellable */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable create(FlowableOnSubscribe source, BackpressureStrategy mode) { ObjectHelper.requireNonNull(source, "source is null"); ObjectHelper.requireNonNull(mode, "mode is null"); return RxJavaPlugins.onAssembly(new FlowableCreate(source, mode)); } /** * Returns a Flowable that calls a Publisher factory to create a Publisher for each new Subscriber * that subscribes. That is, for each subscriber, the actual Publisher that subscriber observes is * determined by the factory function. *

* *

* The defer Subscriber allows you to defer or delay emitting items from a Publisher until such time as a * Subscriber subscribes to the Publisher. This allows a {@link Subscriber} to easily obtain updates or a * refreshed version of the sequence. *

*
Backpressure:
*
The operator itself doesn't interfere with backpressure which is determined by the {@code Publisher} * returned by the {@code supplier}.
*
Scheduler:
*
{@code defer} does not operate by default on a particular {@link Scheduler}.
*
* * @param supplier * the Publisher factory function to invoke for each {@link Subscriber} that subscribes to the * resulting Publisher * @param * the type of the items emitted by the Publisher * @return a Flowable whose {@link Subscriber}s' subscriptions trigger an invocation of the given * Publisher factory function * @see ReactiveX operators documentation: Defer */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable defer(Callable> supplier) { ObjectHelper.requireNonNull(supplier, "supplier is null"); return RxJavaPlugins.onAssembly(new FlowableDefer(supplier)); } /** * Returns a Flowable that emits no items to the {@link Subscriber} and immediately invokes its * {@link Subscriber#onComplete onComplete} method. *

* *

*
Backpressure:
*
This source doesn't produce any elements and effectively ignores downstream backpressure.
*
Scheduler:
*
{@code empty} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of the items (ostensibly) emitted by the Publisher * @return a Flowable that emits no items to the {@link Subscriber} but immediately invokes the * {@link Subscriber}'s {@link Subscriber#onComplete() onComplete} method * @see ReactiveX operators documentation: Empty */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) @SuppressWarnings("unchecked") public static Flowable empty() { return RxJavaPlugins.onAssembly((Flowable) FlowableEmpty.INSTANCE); } /** * Returns a Flowable that invokes a {@link Subscriber}'s {@link Subscriber#onError onError} method when the * Subscriber subscribes to it. *

* *

*
Backpressure:
*
This source doesn't produce any elements and effectively ignores downstream backpressure.
*
Scheduler:
*
{@code error} does not operate by default on a particular {@link Scheduler}.
*
* * @param supplier * a Callable factory to return a Throwable for each individual Subscriber * @param * the type of the items (ostensibly) emitted by the Publisher * @return a Flowable that invokes the {@link Subscriber}'s {@link Subscriber#onError onError} method when * the Subscriber subscribes to it * @see ReactiveX operators documentation: Throw */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable error(Callable supplier) { ObjectHelper.requireNonNull(supplier, "supplier is null"); return RxJavaPlugins.onAssembly(new FlowableError(supplier)); } /** * Returns a Flowable that invokes a {@link Subscriber}'s {@link Subscriber#onError onError} method when the * Subscriber subscribes to it. *

* *

*
Backpressure:
*
This source doesn't produce any elements and effectively ignores downstream backpressure.
*
Scheduler:
*
{@code error} does not operate by default on a particular {@link Scheduler}.
*
* * @param throwable * the particular Throwable to pass to {@link Subscriber#onError onError} * @param * the type of the items (ostensibly) emitted by the Publisher * @return a Flowable that invokes the {@link Subscriber}'s {@link Subscriber#onError onError} method when * the Subscriber subscribes to it * @see ReactiveX operators documentation: Throw */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable error(final Throwable throwable) { ObjectHelper.requireNonNull(throwable, "throwable is null"); return error(Functions.justCallable(throwable)); } /** * Converts an Array into a Publisher that emits the items in the Array. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and iterates the given {@code array} * on demand (i.e., when requested).
*
Scheduler:
*
{@code fromArray} does not operate by default on a particular {@link Scheduler}.
*
* * @param items * the array of elements * @param * the type of items in the Array and the type of items to be emitted by the resulting Publisher * @return a Flowable that emits each item in the source Array * @see ReactiveX operators documentation: From */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable fromArray(T... items) { ObjectHelper.requireNonNull(items, "items is null"); if (items.length == 0) { return empty(); } if (items.length == 1) { return just(items[0]); } return RxJavaPlugins.onAssembly(new FlowableFromArray(items)); } /** * Returns a Flowable that, when a Subscriber subscribes to it, invokes a function you specify and then * emits the value returned from that function. *

* *

* This allows you to defer the execution of the function you specify until a Subscriber subscribes to the * Publisher. That is to say, it makes the function "lazy." *

*
Backpressure:
*
The operator honors backpressure from downstream.
*
Scheduler:
*
{@code fromCallable} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If the {@link Callable} throws an exception, the respective {@link Throwable} is * delivered to the downstream via {@link Subscriber#onError(Throwable)}, * except when the downstream has canceled this {@code Flowable} source. * In this latter case, the {@code Throwable} is delivered to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} as an {@link io.reactivex.exceptions.UndeliverableException UndeliverableException}. *
*
* * @param supplier * a function, the execution of which should be deferred; {@code fromCallable} will invoke this * function only when a Subscriber subscribes to the Publisher that {@code fromCallable} returns * @param * the type of the item emitted by the Publisher * @return a Flowable whose {@link Subscriber}s' subscriptions trigger an invocation of the given function * @see #defer(Callable) * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable fromCallable(Callable supplier) { ObjectHelper.requireNonNull(supplier, "supplier is null"); return RxJavaPlugins.onAssembly(new FlowableFromCallable(supplier)); } /** * Converts a {@link Future} into a Publisher. *

* *

* You can convert any object that supports the {@link Future} interface into a Publisher that emits the * return value of the {@link Future#get} method of that object by passing the object into the {@code from} * method. *

* Important note: This Publisher is blocking on the thread it gets subscribed on; you cannot cancel it. *

* Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the * cancellation effect: {@code futurePublisher.doOnCancel(() -> future.cancel(true));}. *

*
Backpressure:
*
The operator honors backpressure from downstream.
*
Scheduler:
*
{@code fromFuture} does not operate by default on a particular {@link Scheduler}.
*
* * @param future * the source {@link Future} * @param * the type of object that the {@link Future} returns, and also the type of item to be emitted by * the resulting Publisher * @return a Flowable that emits the item from the source {@link Future} * @see ReactiveX operators documentation: From */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable fromFuture(Future future) { ObjectHelper.requireNonNull(future, "future is null"); return RxJavaPlugins.onAssembly(new FlowableFromFuture(future, 0L, null)); } /** * Converts a {@link Future} into a Publisher, with a timeout on the Future. *

* *

* You can convert any object that supports the {@link Future} interface into a Publisher that emits the * return value of the {@link Future#get} method of that object by passing the object into the {@code fromFuture} * method. *

* Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the * cancellation effect: {@code futurePublisher.doOnCancel(() -> future.cancel(true));}. *

* Important note: This Publisher is blocking on the thread it gets subscribed on; you cannot cancel it. *

*
Backpressure:
*
The operator honors backpressure from downstream.
*
Scheduler:
*
{@code fromFuture} does not operate by default on a particular {@link Scheduler}.
*
* * @param future * the source {@link Future} * @param timeout * the maximum time to wait before calling {@code get} * @param unit * the {@link TimeUnit} of the {@code timeout} argument * @param * the type of object that the {@link Future} returns, and also the type of item to be emitted by * the resulting Publisher * @return a Flowable that emits the item from the source {@link Future} * @see ReactiveX operators documentation: From */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable fromFuture(Future future, long timeout, TimeUnit unit) { ObjectHelper.requireNonNull(future, "future is null"); ObjectHelper.requireNonNull(unit, "unit is null"); return RxJavaPlugins.onAssembly(new FlowableFromFuture(future, timeout, unit)); } /** * Converts a {@link Future} into a Publisher, with a timeout on the Future. *

* *

* You can convert any object that supports the {@link Future} interface into a Publisher that emits the * return value of the {@link Future#get} method of that object by passing the object into the {@code from} * method. *

* Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the * cancellation effect: {@code futurePublisher.doOnCancel(() -> future.cancel(true));}. *

* Important note: This Publisher is blocking; you cannot cancel it. *

*
Backpressure:
*
The operator honors backpressure from downstream.
*
Scheduler:
*
{@code fromFuture} does not operate by default on a particular {@link Scheduler}.
*
* * @param future * the source {@link Future} * @param timeout * the maximum time to wait before calling {@code get} * @param unit * the {@link TimeUnit} of the {@code timeout} argument * @param scheduler * the {@link Scheduler} to wait for the Future on. Use a Scheduler such as * {@link Schedulers#io()} that can block and wait on the Future * @param * the type of object that the {@link Future} returns, and also the type of item to be emitted by * the resulting Publisher * @return a Flowable that emits the item from the source {@link Future} * @see ReactiveX operators documentation: From */ @SuppressWarnings({ "unchecked", "cast" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public static Flowable fromFuture(Future future, long timeout, TimeUnit unit, Scheduler scheduler) { ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return fromFuture((Future)future, timeout, unit).subscribeOn(scheduler); } /** * Converts a {@link Future}, operating on a specified {@link Scheduler}, into a Publisher. *

* *

* You can convert any object that supports the {@link Future} interface into a Publisher that emits the * return value of the {@link Future#get} method of that object by passing the object into the {@code from} * method. *

* Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the * cancellation effect: {@code futurePublisher.doOnCancel(() -> future.cancel(true));}. *

*
Backpressure:
*
The operator honors backpressure from downstream.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param future * the source {@link Future} * @param scheduler * the {@link Scheduler} to wait for the Future on. Use a Scheduler such as * {@link Schedulers#io()} that can block and wait on the Future * @param * the type of object that the {@link Future} returns, and also the type of item to be emitted by * the resulting Publisher * @return a Flowable that emits the item from the source {@link Future} * @see ReactiveX operators documentation: From */ @SuppressWarnings({ "cast", "unchecked" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public static Flowable fromFuture(Future future, Scheduler scheduler) { ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return fromFuture((Future)future).subscribeOn(scheduler); } /** * Converts an {@link Iterable} sequence into a Publisher that emits the items in the sequence. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and iterates the given {@code iterable} * on demand (i.e., when requested).
*
Scheduler:
*
{@code fromIterable} does not operate by default on a particular {@link Scheduler}.
*
* * @param source * the source {@link Iterable} sequence * @param * the type of items in the {@link Iterable} sequence and the type of items to be emitted by the * resulting Publisher * @return a Flowable that emits each item in the source {@link Iterable} sequence * @see ReactiveX operators documentation: From */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable fromIterable(Iterable source) { ObjectHelper.requireNonNull(source, "source is null"); return RxJavaPlugins.onAssembly(new FlowableFromIterable(source)); } /** * Converts an arbitrary Reactive Streams Publisher into a Flowable if not already a * Flowable. *

* The {@link Publisher} must follow the * Reactive Streams specification. * Violating the specification may result in undefined behavior. *

* If possible, use {@link #create(FlowableOnSubscribe, BackpressureStrategy)} to create a * source-like {@code Flowable} instead. *

* Note that even though {@link Publisher} appears to be a functional interface, it * is not recommended to implement it through a lambda as the specification requires * state management that is not achievable with a stateless lambda. *

*
Backpressure:
*
The operator is a pass-through for backpressure and its behavior is determined by the * backpressure behavior of the wrapped publisher.
*
Scheduler:
*
{@code fromPublisher} does not operate by default on a particular {@link Scheduler}.
*
* @param the value type of the flow * @param source the Publisher to convert * @return the new Flowable instance * @throws NullPointerException if the {@code source} {@code Publisher} is null * @see #create(FlowableOnSubscribe, BackpressureStrategy) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) @SuppressWarnings("unchecked") public static Flowable fromPublisher(final Publisher source) { if (source instanceof Flowable) { return RxJavaPlugins.onAssembly((Flowable)source); } ObjectHelper.requireNonNull(source, "source is null"); return RxJavaPlugins.onAssembly(new FlowableFromPublisher(source)); } /** * Returns a cold, synchronous, stateless and backpressure-aware generator of values. *

* Note that the {@link Emitter#onNext}, {@link Emitter#onError} and * {@link Emitter#onComplete} methods provided to the function via the {@link Emitter} instance should be called synchronously, * never concurrently and only while the function body is executing. Calling them from multiple threads * or outside the function call is not supported and leads to an undefined behavior. *

*
Backpressure:
*
The operator honors downstream backpressure.
*
Scheduler:
*
{@code generate} does not operate by default on a particular {@link Scheduler}.
*
* * @param the generated value type * @param generator the Consumer called whenever a particular downstream Subscriber has * requested a value. The callback then should call {@code onNext}, {@code onError} or * {@code onComplete} to signal a value or a terminal event. Signaling multiple {@code onNext} * in a call will make the operator signal {@code IllegalStateException}. * @return the new Flowable instance */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable generate(final Consumer> generator) { ObjectHelper.requireNonNull(generator, "generator is null"); return generate(Functions.nullSupplier(), FlowableInternalHelper.simpleGenerator(generator), Functions.emptyConsumer()); } /** * Returns a cold, synchronous, stateful and backpressure-aware generator of values. *

* Note that the {@link Emitter#onNext}, {@link Emitter#onError} and * {@link Emitter#onComplete} methods provided to the function via the {@link Emitter} instance should be called synchronously, * never concurrently and only while the function body is executing. Calling them from multiple threads * or outside the function call is not supported and leads to an undefined behavior. *

*
Backpressure:
*
The operator honors downstream backpressure.
*
Scheduler:
*
{@code generate} does not operate by default on a particular {@link Scheduler}.
*
* * @param the type of the per-Subscriber state * @param the generated value type * @param initialState the Callable to generate the initial state for each Subscriber * @param generator the Consumer called with the current state whenever a particular downstream Subscriber has * requested a value. The callback then should call {@code onNext}, {@code onError} or * {@code onComplete} to signal a value or a terminal event. Signaling multiple {@code onNext} * in a call will make the operator signal {@code IllegalStateException}. * @return the new Flowable instance */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable generate(Callable initialState, final BiConsumer> generator) { ObjectHelper.requireNonNull(generator, "generator is null"); return generate(initialState, FlowableInternalHelper.simpleBiGenerator(generator), Functions.emptyConsumer()); } /** * Returns a cold, synchronous, stateful and backpressure-aware generator of values. *

* Note that the {@link Emitter#onNext}, {@link Emitter#onError} and * {@link Emitter#onComplete} methods provided to the function via the {@link Emitter} instance should be called synchronously, * never concurrently and only while the function body is executing. Calling them from multiple threads * or outside the function call is not supported and leads to an undefined behavior. *

*
Backpressure:
*
The operator honors downstream backpressure.
*
Scheduler:
*
{@code generate} does not operate by default on a particular {@link Scheduler}.
*
* * @param the type of the per-Subscriber state * @param the generated value type * @param initialState the Callable to generate the initial state for each Subscriber * @param generator the Consumer called with the current state whenever a particular downstream Subscriber has * requested a value. The callback then should call {@code onNext}, {@code onError} or * {@code onComplete} to signal a value or a terminal event. Signaling multiple {@code onNext} * in a call will make the operator signal {@code IllegalStateException}. * @param disposeState the Consumer that is called with the current state when the generator * terminates the sequence or it gets canceled * @return the new Flowable instance */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable generate(Callable initialState, final BiConsumer> generator, Consumer disposeState) { ObjectHelper.requireNonNull(generator, "generator is null"); return generate(initialState, FlowableInternalHelper.simpleBiGenerator(generator), disposeState); } /** * Returns a cold, synchronous, stateful and backpressure-aware generator of values. *

* Note that the {@link Emitter#onNext}, {@link Emitter#onError} and * {@link Emitter#onComplete} methods provided to the function via the {@link Emitter} instance should be called synchronously, * never concurrently and only while the function body is executing. Calling them from multiple threads * or outside the function call is not supported and leads to an undefined behavior. *

*
Backpressure:
*
The operator honors downstream backpressure.
*
Scheduler:
*
{@code generate} does not operate by default on a particular {@link Scheduler}.
*
* * @param the type of the per-Subscriber state * @param the generated value type * @param initialState the Callable to generate the initial state for each Subscriber * @param generator the Function called with the current state whenever a particular downstream Subscriber has * requested a value. The callback then should call {@code onNext}, {@code onError} or * {@code onComplete} to signal a value or a terminal event and should return a (new) state for * the next invocation. Signaling multiple {@code onNext} * in a call will make the operator signal {@code IllegalStateException}. * @return the new Flowable instance */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable generate(Callable initialState, BiFunction, S> generator) { return generate(initialState, generator, Functions.emptyConsumer()); } /** * Returns a cold, synchronous, stateful and backpressure-aware generator of values. *

* Note that the {@link Emitter#onNext}, {@link Emitter#onError} and * {@link Emitter#onComplete} methods provided to the function via the {@link Emitter} instance should be called synchronously, * never concurrently and only while the function body is executing. Calling them from multiple threads * or outside the function call is not supported and leads to an undefined behavior. *

*
Backpressure:
*
The operator honors downstream backpressure.
*
Scheduler:
*
{@code generate} does not operate by default on a particular {@link Scheduler}.
*
* * @param the type of the per-Subscriber state * @param the generated value type * @param initialState the Callable to generate the initial state for each Subscriber * @param generator the Function called with the current state whenever a particular downstream Subscriber has * requested a value. The callback then should call {@code onNext}, {@code onError} or * {@code onComplete} to signal a value or a terminal event and should return a (new) state for * the next invocation. Signaling multiple {@code onNext} * in a call will make the operator signal {@code IllegalStateException}. * @param disposeState the Consumer that is called with the current state when the generator * terminates the sequence or it gets canceled * @return the new Flowable instance */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable generate(Callable initialState, BiFunction, S> generator, Consumer disposeState) { ObjectHelper.requireNonNull(initialState, "initialState is null"); ObjectHelper.requireNonNull(generator, "generator is null"); ObjectHelper.requireNonNull(disposeState, "disposeState is null"); return RxJavaPlugins.onAssembly(new FlowableGenerate(initialState, generator, disposeState)); } /** * Returns a Flowable that emits a {@code 0L} after the {@code initialDelay} and ever-increasing numbers * after each {@code period} of time thereafter. *

* *

*
Backpressure:
*
The operator generates values based on time and ignores downstream backpressure which * may lead to {@code MissingBackpressureException} at some point in the chain. * Consumers should consider applying one of the {@code onBackpressureXXX} operators as well.
*
Scheduler:
*
{@code interval} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param initialDelay * the initial delay time to wait before emitting the first value of 0L * @param period * the period of time between emissions of the subsequent numbers * @param unit * the time unit for both {@code initialDelay} and {@code period} * @return a Flowable that emits a 0L after the {@code initialDelay} and ever-increasing numbers after * each {@code period} of time thereafter * @see ReactiveX operators documentation: Interval * @since 1.0.12 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public static Flowable interval(long initialDelay, long period, TimeUnit unit) { return interval(initialDelay, period, unit, Schedulers.computation()); } /** * Returns a Flowable that emits a {@code 0L} after the {@code initialDelay} and ever-increasing numbers * after each {@code period} of time thereafter, on a specified {@link Scheduler}. *

* *

*
Backpressure:
*
The operator generates values based on time and ignores downstream backpressure which * may lead to {@code MissingBackpressureException} at some point in the chain. * Consumers should consider applying one of the {@code onBackpressureXXX} operators as well.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param initialDelay * the initial delay time to wait before emitting the first value of 0L * @param period * the period of time between emissions of the subsequent numbers * @param unit * the time unit for both {@code initialDelay} and {@code period} * @param scheduler * the Scheduler on which the waiting happens and items are emitted * @return a Flowable that emits a 0L after the {@code initialDelay} and ever-increasing numbers after * each {@code period} of time thereafter, while running on the given Scheduler * @see ReactiveX operators documentation: Interval * @since 1.0.12 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public static Flowable interval(long initialDelay, long period, TimeUnit unit, Scheduler scheduler) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableInterval(Math.max(0L, initialDelay), Math.max(0L, period), unit, scheduler)); } /** * Returns a Flowable that emits a sequential number every specified interval of time. *

* *

*
Backpressure:
*
The operator signals a {@code MissingBackpressureException} if the downstream * is not ready to receive the next value.
*
Scheduler:
*
{@code interval} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param period * the period size in time units (see below) * @param unit * time units to use for the interval size * @return a Flowable that emits a sequential number each time interval * @see ReactiveX operators documentation: Interval */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public static Flowable interval(long period, TimeUnit unit) { return interval(period, period, unit, Schedulers.computation()); } /** * Returns a Flowable that emits a sequential number every specified interval of time, on a * specified Scheduler. *

* *

*
Backpressure:
*
The operator generates values based on time and ignores downstream backpressure which * may lead to {@code MissingBackpressureException} at some point in the chain. * Consumers should consider applying one of the {@code onBackpressureXXX} operators as well.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param period * the period size in time units (see below) * @param unit * time units to use for the interval size * @param scheduler * the Scheduler to use for scheduling the items * @return a Flowable that emits a sequential number each time interval * @see ReactiveX operators documentation: Interval */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public static Flowable interval(long period, TimeUnit unit, Scheduler scheduler) { return interval(period, period, unit, scheduler); } /** * Signals a range of long values, the first after some initial delay and the rest periodically after. *

* The sequence completes immediately after the last value (start + count - 1) has been reached. *

*
Backpressure:
*
The operator signals a {@code MissingBackpressureException} if the downstream can't keep up.
*
Scheduler:
*
{@code intervalRange} by default operates on the {@link Schedulers#computation() computation} {@link Scheduler}.
*
* @param start that start value of the range * @param count the number of values to emit in total, if zero, the operator emits an onComplete after the initial delay. * @param initialDelay the initial delay before signaling the first value (the start) * @param period the period between subsequent values * @param unit the unit of measure of the initialDelay and period amounts * @return the new Flowable instance */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public static Flowable intervalRange(long start, long count, long initialDelay, long period, TimeUnit unit) { return intervalRange(start, count, initialDelay, period, unit, Schedulers.computation()); } /** * Signals a range of long values, the first after some initial delay and the rest periodically after. *

* The sequence completes immediately after the last value (start + count - 1) has been reached. *

*
Backpressure:
*
The operator signals a {@code MissingBackpressureException} if the downstream can't keep up.
*
Scheduler:
*
you provide the {@link Scheduler}.
*
* @param start that start value of the range * @param count the number of values to emit in total, if zero, the operator emits an onComplete after the initial delay. * @param initialDelay the initial delay before signaling the first value (the start) * @param period the period between subsequent values * @param unit the unit of measure of the initialDelay and period amounts * @param scheduler the target scheduler where the values and terminal signals will be emitted * @return the new Flowable instance */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public static Flowable intervalRange(long start, long count, long initialDelay, long period, TimeUnit unit, Scheduler scheduler) { if (count < 0L) { throw new IllegalArgumentException("count >= 0 required but it was " + count); } if (count == 0L) { return Flowable.empty().delay(initialDelay, unit, scheduler); } long end = start + (count - 1); if (start > 0 && end < 0) { throw new IllegalArgumentException("Overflow! start + count is bigger than Long.MAX_VALUE"); } ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableIntervalRange(start, end, Math.max(0L, initialDelay), Math.max(0L, period), unit, scheduler)); } /** * Returns a Flowable that signals the given (constant reference) item and then completes. *

* *

* Note that the item is taken and re-emitted as is and not computed by any means by {@code just}. Use {@link #fromCallable(Callable)} * to generate a single item on demand (when {@code Subscriber}s subscribe to it). *

* See the multi-parameter overloads of {@code just} to emit more than one (constant reference) items one after the other. * Use {@link #fromArray(Object...)} to emit an arbitrary number of items that are known upfront. *

* To emit the items of an {@link Iterable} sequence (such as a {@link java.util.List}), use {@link #fromIterable(Iterable)}. *

*
Backpressure:
*
The operator honors backpressure from downstream.
*
Scheduler:
*
{@code just} does not operate by default on a particular {@link Scheduler}.
*
* * @param item * the item to emit * @param * the type of that item * @return a Flowable that emits {@code value} as a single item and then completes * @see ReactiveX operators documentation: Just * @see #just(Object, Object) * @see #fromCallable(Callable) * @see #fromArray(Object...) * @see #fromIterable(Iterable) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable just(T item) { ObjectHelper.requireNonNull(item, "item is null"); return RxJavaPlugins.onAssembly(new FlowableJust(item)); } /** * Converts two items into a Publisher that emits those items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
*
Scheduler:
*
{@code just} does not operate by default on a particular {@link Scheduler}.
*
* * @param item1 * first item * @param item2 * second item * @param * the type of these items * @return a Flowable that emits each item * @see ReactiveX operators documentation: Just */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable just(T item1, T item2) { ObjectHelper.requireNonNull(item1, "item1 is null"); ObjectHelper.requireNonNull(item2, "item2 is null"); return fromArray(item1, item2); } /** * Converts three items into a Publisher that emits those items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
*
Scheduler:
*
{@code just} does not operate by default on a particular {@link Scheduler}.
*
* * @param item1 * first item * @param item2 * second item * @param item3 * third item * @param * the type of these items * @return a Flowable that emits each item * @see ReactiveX operators documentation: Just */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable just(T item1, T item2, T item3) { ObjectHelper.requireNonNull(item1, "item1 is null"); ObjectHelper.requireNonNull(item2, "item2 is null"); ObjectHelper.requireNonNull(item3, "item3 is null"); return fromArray(item1, item2, item3); } /** * Converts four items into a Publisher that emits those items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
*
Scheduler:
*
{@code just} does not operate by default on a particular {@link Scheduler}.
*
* * @param item1 * first item * @param item2 * second item * @param item3 * third item * @param item4 * fourth item * @param * the type of these items * @return a Flowable that emits each item * @see ReactiveX operators documentation: Just */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable just(T item1, T item2, T item3, T item4) { ObjectHelper.requireNonNull(item1, "item1 is null"); ObjectHelper.requireNonNull(item2, "item2 is null"); ObjectHelper.requireNonNull(item3, "item3 is null"); ObjectHelper.requireNonNull(item4, "item4 is null"); return fromArray(item1, item2, item3, item4); } /** * Converts five items into a Publisher that emits those items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
*
Scheduler:
*
{@code just} does not operate by default on a particular {@link Scheduler}.
*
* * @param item1 * first item * @param item2 * second item * @param item3 * third item * @param item4 * fourth item * @param item5 * fifth item * @param * the type of these items * @return a Flowable that emits each item * @see ReactiveX operators documentation: Just */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable just(T item1, T item2, T item3, T item4, T item5) { ObjectHelper.requireNonNull(item1, "item1 is null"); ObjectHelper.requireNonNull(item2, "item2 is null"); ObjectHelper.requireNonNull(item3, "item3 is null"); ObjectHelper.requireNonNull(item4, "item4 is null"); ObjectHelper.requireNonNull(item5, "item5 is null"); return fromArray(item1, item2, item3, item4, item5); } /** * Converts six items into a Publisher that emits those items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
*
Scheduler:
*
{@code just} does not operate by default on a particular {@link Scheduler}.
*
* * @param item1 * first item * @param item2 * second item * @param item3 * third item * @param item4 * fourth item * @param item5 * fifth item * @param item6 * sixth item * @param * the type of these items * @return a Flowable that emits each item * @see ReactiveX operators documentation: Just */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable just(T item1, T item2, T item3, T item4, T item5, T item6) { ObjectHelper.requireNonNull(item1, "item1 is null"); ObjectHelper.requireNonNull(item2, "item2 is null"); ObjectHelper.requireNonNull(item3, "item3 is null"); ObjectHelper.requireNonNull(item4, "item4 is null"); ObjectHelper.requireNonNull(item5, "item5 is null"); ObjectHelper.requireNonNull(item6, "item6 is null"); return fromArray(item1, item2, item3, item4, item5, item6); } /** * Converts seven items into a Publisher that emits those items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
*
Scheduler:
*
{@code just} does not operate by default on a particular {@link Scheduler}.
*
* * @param item1 * first item * @param item2 * second item * @param item3 * third item * @param item4 * fourth item * @param item5 * fifth item * @param item6 * sixth item * @param item7 * seventh item * @param * the type of these items * @return a Flowable that emits each item * @see ReactiveX operators documentation: Just */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable just(T item1, T item2, T item3, T item4, T item5, T item6, T item7) { ObjectHelper.requireNonNull(item1, "item1 is null"); ObjectHelper.requireNonNull(item2, "item2 is null"); ObjectHelper.requireNonNull(item3, "item3 is null"); ObjectHelper.requireNonNull(item4, "item4 is null"); ObjectHelper.requireNonNull(item5, "item5 is null"); ObjectHelper.requireNonNull(item6, "item6 is null"); ObjectHelper.requireNonNull(item7, "item7 is null"); return fromArray(item1, item2, item3, item4, item5, item6, item7); } /** * Converts eight items into a Publisher that emits those items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
*
Scheduler:
*
{@code just} does not operate by default on a particular {@link Scheduler}.
*
* * @param item1 * first item * @param item2 * second item * @param item3 * third item * @param item4 * fourth item * @param item5 * fifth item * @param item6 * sixth item * @param item7 * seventh item * @param item8 * eighth item * @param * the type of these items * @return a Flowable that emits each item * @see ReactiveX operators documentation: Just */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable just(T item1, T item2, T item3, T item4, T item5, T item6, T item7, T item8) { ObjectHelper.requireNonNull(item1, "item1 is null"); ObjectHelper.requireNonNull(item2, "item2 is null"); ObjectHelper.requireNonNull(item3, "item3 is null"); ObjectHelper.requireNonNull(item4, "item4 is null"); ObjectHelper.requireNonNull(item5, "item5 is null"); ObjectHelper.requireNonNull(item6, "item6 is null"); ObjectHelper.requireNonNull(item7, "item7 is null"); ObjectHelper.requireNonNull(item8, "item8 is null"); return fromArray(item1, item2, item3, item4, item5, item6, item7, item8); } /** * Converts nine items into a Publisher that emits those items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
*
Scheduler:
*
{@code just} does not operate by default on a particular {@link Scheduler}.
*
* * @param item1 * first item * @param item2 * second item * @param item3 * third item * @param item4 * fourth item * @param item5 * fifth item * @param item6 * sixth item * @param item7 * seventh item * @param item8 * eighth item * @param item9 * ninth item * @param * the type of these items * @return a Flowable that emits each item * @see ReactiveX operators documentation: Just */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable just(T item1, T item2, T item3, T item4, T item5, T item6, T item7, T item8, T item9) { ObjectHelper.requireNonNull(item1, "item1 is null"); ObjectHelper.requireNonNull(item2, "item2 is null"); ObjectHelper.requireNonNull(item3, "item3 is null"); ObjectHelper.requireNonNull(item4, "item4 is null"); ObjectHelper.requireNonNull(item5, "item5 is null"); ObjectHelper.requireNonNull(item6, "item6 is null"); ObjectHelper.requireNonNull(item7, "item7 is null"); ObjectHelper.requireNonNull(item8, "item8 is null"); ObjectHelper.requireNonNull(item9, "item9 is null"); return fromArray(item1, item2, item3, item4, item5, item6, item7, item8, item9); } /** * Converts ten items into a Publisher that emits those items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
*
Scheduler:
*
{@code just} does not operate by default on a particular {@link Scheduler}.
*
* * @param item1 * first item * @param item2 * second item * @param item3 * third item * @param item4 * fourth item * @param item5 * fifth item * @param item6 * sixth item * @param item7 * seventh item * @param item8 * eighth item * @param item9 * ninth item * @param item10 * tenth item * @param * the type of these items * @return a Flowable that emits each item * @see ReactiveX operators documentation: Just */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable just(T item1, T item2, T item3, T item4, T item5, T item6, T item7, T item8, T item9, T item10) { ObjectHelper.requireNonNull(item1, "item1 is null"); ObjectHelper.requireNonNull(item2, "item2 is null"); ObjectHelper.requireNonNull(item3, "item3 is null"); ObjectHelper.requireNonNull(item4, "item4 is null"); ObjectHelper.requireNonNull(item5, "item5 is null"); ObjectHelper.requireNonNull(item6, "item6 is null"); ObjectHelper.requireNonNull(item7, "item7 is null"); ObjectHelper.requireNonNull(item8, "item8 is null"); ObjectHelper.requireNonNull(item9, "item9 is null"); ObjectHelper.requireNonNull(item10, "item10 is null"); return fromArray(item1, item2, item3, item4, item5, item6, item7, item8, item9, item10); } /** * Flattens an Iterable of Publishers into one Publisher, without any transformation, while limiting the * number of concurrent subscriptions to these Publishers. *

* *

* You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code merge} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code merge} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting * {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled. * If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the * first one's error or, depending on the concurrency of the sources, may terminate with a * {@code CompositeException} containing two or more of the various error signals. * {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s * signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a * (composite) error will be sent to the same global error handler. * Use {@link #mergeDelayError(Iterable, int, int)} to merge sources and terminate only when all source {@code Publisher}s * have completed or failed with an error. *
*
* * @param the common element base type * @param sources * the Iterable of Publishers * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @param bufferSize * the number of items to prefetch from each inner Publisher * @return a Flowable that emits items that are the result of flattening the items emitted by the * Publishers in the Iterable * @throws IllegalArgumentException * if {@code maxConcurrency} is less than or equal to 0 * @see ReactiveX operators documentation: Merge * @see #mergeDelayError(Iterable, int, int) */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable merge(Iterable> sources, int maxConcurrency, int bufferSize) { return fromIterable(sources).flatMap((Function)Functions.identity(), false, maxConcurrency, bufferSize); } /** * Flattens an Iterable of Publishers into one Publisher, without any transformation, while limiting the * number of concurrent subscriptions to these Publishers. *

* *

* You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code merge} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeArray} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting * {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled. * If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the * first one's error or, depending on the concurrency of the sources, may terminate with a * {@code CompositeException} containing two or more of the various error signals. * {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s * signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a * (composite) error will be sent to the same global error handler. * Use {@link #mergeArrayDelayError(int, int, Publisher[])} to merge sources and terminate only when all source {@code Publisher}s * have completed or failed with an error. *
*
* * @param the common element base type * @param sources * the array of Publishers * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @param bufferSize * the number of items to prefetch from each inner Publisher * @return a Flowable that emits items that are the result of flattening the items emitted by the * Publishers in the Iterable * @throws IllegalArgumentException * if {@code maxConcurrency} is less than or equal to 0 * @see ReactiveX operators documentation: Merge * @see #mergeArrayDelayError(int, int, Publisher...) */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeArray(int maxConcurrency, int bufferSize, Publisher... sources) { return fromArray(sources).flatMap((Function)Functions.identity(), false, maxConcurrency, bufferSize); } /** * Flattens an Iterable of Publishers into one Publisher, without any transformation. *

* *

* You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code merge} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code merge} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting * {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled. * If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the * first one's error or, depending on the concurrency of the sources, may terminate with a * {@code CompositeException} containing two or more of the various error signals. * {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s * signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a * (composite) error will be sent to the same global error handler. * Use {@link #mergeDelayError(Iterable)} to merge sources and terminate only when all source {@code Publisher}s * have completed or failed with an error. *
*
* * @param the common element base type * @param sources * the Iterable of Publishers * @return a Flowable that emits items that are the result of flattening the items emitted by the * Publishers in the Iterable * @see ReactiveX operators documentation: Merge * @see #mergeDelayError(Iterable) */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable merge(Iterable> sources) { return fromIterable(sources).flatMap((Function)Functions.identity()); } /** * Flattens an Iterable of Publishers into one Publisher, without any transformation, while limiting the * number of concurrent subscriptions to these Publishers. *

* *

* You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code merge} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code merge} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting * {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled. * If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the * first one's error or, depending on the concurrency of the sources, may terminate with a * {@code CompositeException} containing two or more of the various error signals. * {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s * signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a * (composite) error will be sent to the same global error handler. * Use {@link #mergeDelayError(Iterable, int)} to merge sources and terminate only when all source {@code Publisher}s * have completed or failed with an error. *
*
* * @param the common element base type * @param sources * the Iterable of Publishers * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @return a Flowable that emits items that are the result of flattening the items emitted by the * Publishers in the Iterable * @throws IllegalArgumentException * if {@code maxConcurrency} is less than or equal to 0 * @see ReactiveX operators documentation: Merge * @see #mergeDelayError(Iterable, int) */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable merge(Iterable> sources, int maxConcurrency) { return fromIterable(sources).flatMap((Function)Functions.identity(), maxConcurrency); } /** * Flattens a Publisher that emits Publishers into a single Publisher that emits the items emitted by * those Publishers, without any transformation. *

* *

* You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code merge} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. The outer {@code Publisher} is consumed * in unbounded mode (i.e., no backpressure is applied to it). The inner {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code merge} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting * {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled. * If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the * first one's error or, depending on the concurrency of the sources, may terminate with a * {@code CompositeException} containing two or more of the various error signals. * {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s * signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a * (composite) error will be sent to the same global error handler. * Use {@link #mergeDelayError(Publisher)} to merge sources and terminate only when all source {@code Publisher}s * have completed or failed with an error. *
*
* * @param the common element base type * @param sources * a Publisher that emits Publishers * @return a Flowable that emits items that are the result of flattening the Publishers emitted by the * {@code source} Publisher * @see ReactiveX operators documentation: Merge * @see #mergeDelayError(Publisher) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable merge(Publisher> sources) { return merge(sources, bufferSize()); } /** * Flattens a Publisher that emits Publishers into a single Publisher that emits the items emitted by * those Publishers, without any transformation, while limiting the maximum number of concurrent * subscriptions to these Publishers. *

* *

* You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code merge} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. Both the outer and inner {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code merge} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting * {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled. * If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the * first one's error or, depending on the concurrency of the sources, may terminate with a * {@code CompositeException} containing two or more of the various error signals. * {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s * signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a * (composite) error will be sent to the same global error handler. * Use {@link #mergeDelayError(Publisher, int)} to merge sources and terminate only when all source {@code Publisher}s * have completed or failed with an error. *
*
* * @param the common element base type * @param sources * a Publisher that emits Publishers * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @return a Flowable that emits items that are the result of flattening the Publishers emitted by the * {@code source} Publisher * @throws IllegalArgumentException * if {@code maxConcurrency} is less than or equal to 0 * @see ReactiveX operators documentation: Merge * @see #mergeDelayError(Publisher, int) * @since 1.1.0 */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable merge(Publisher> sources, int maxConcurrency) { return fromPublisher(sources).flatMap((Function)Functions.identity(), maxConcurrency); } /** * Flattens an Array of Publishers into one Publisher, without any transformation. *

* *

* You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code merge} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeArray} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting * {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled. * If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the * first one's error or, depending on the concurrency of the sources, may terminate with a * {@code CompositeException} containing two or more of the various error signals. * {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s * signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a * (composite) error will be sent to the same global error handler. * Use {@link #mergeArrayDelayError(Publisher...)} to merge sources and terminate only when all source {@code Publisher}s * have completed or failed with an error. *
*
* * @param the common element base type * @param sources * the array of Publishers * @return a Flowable that emits all of the items emitted by the Publishers in the Array * @see ReactiveX operators documentation: Merge * @see #mergeArrayDelayError(Publisher...) */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeArray(Publisher... sources) { return fromArray(sources).flatMap((Function)Functions.identity(), sources.length); } /** * Flattens two Publishers into a single Publisher, without any transformation. *

* *

* You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code merge} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code merge} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting * {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled. * If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the * first one's error or, depending on the concurrency of the sources, may terminate with a * {@code CompositeException} containing two or more of the various error signals. * {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s * signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a * (composite) error will be sent to the same global error handler. * Use {@link #mergeDelayError(Publisher, Publisher)} to merge sources and terminate only when all source {@code Publisher}s * have completed or failed with an error. *
*
* * @param the common element base type * @param source1 * a Publisher to be merged * @param source2 * a Publisher to be merged * @return a Flowable that emits all of the items emitted by the source Publishers * @see ReactiveX operators documentation: Merge * @see #mergeDelayError(Publisher, Publisher) */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable merge(Publisher source1, Publisher source2) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); return fromArray(source1, source2).flatMap((Function)Functions.identity(), false, 2); } /** * Flattens three Publishers into a single Publisher, without any transformation. *

* *

* You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code merge} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code merge} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting * {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled. * If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the * first one's error or, depending on the concurrency of the sources, may terminate with a * {@code CompositeException} containing two or more of the various error signals. * {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s * signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a * (composite) error will be sent to the same global error handler. * Use {@link #mergeDelayError(Publisher, Publisher, Publisher)} to merge sources and terminate only when all source {@code Publisher}s * have completed or failed with an error. *
*
* * @param the common element base type * @param source1 * a Publisher to be merged * @param source2 * a Publisher to be merged * @param source3 * a Publisher to be merged * @return a Flowable that emits all of the items emitted by the source Publishers * @see ReactiveX operators documentation: Merge * @see #mergeDelayError(Publisher, Publisher, Publisher) */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable merge(Publisher source1, Publisher source2, Publisher source3) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); return fromArray(source1, source2, source3).flatMap((Function)Functions.identity(), false, 3); } /** * Flattens four Publishers into a single Publisher, without any transformation. *

* *

* You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code merge} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code merge} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting * {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled. * If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the * first one's error or, depending on the concurrency of the sources, may terminate with a * {@code CompositeException} containing two or more of the various error signals. * {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s * signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a * (composite) error will be sent to the same global error handler. * Use {@link #mergeDelayError(Publisher, Publisher, Publisher, Publisher)} to merge sources and terminate only when all source {@code Publisher}s * have completed or failed with an error. *
*
* * @param the common element base type * @param source1 * a Publisher to be merged * @param source2 * a Publisher to be merged * @param source3 * a Publisher to be merged * @param source4 * a Publisher to be merged * @return a Flowable that emits all of the items emitted by the source Publishers * @see ReactiveX operators documentation: Merge * @see #mergeDelayError(Publisher, Publisher, Publisher, Publisher) */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable merge( Publisher source1, Publisher source2, Publisher source3, Publisher source4) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); return fromArray(source1, source2, source3, source4).flatMap((Function)Functions.identity(), false, 4); } /** * Flattens an Iterable of Publishers into one Publisher, in a way that allows a Subscriber to receive all * successfully emitted items from each of the source Publishers without being interrupted by an error * notification from one of them. *

* This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that * error notification until all of the merged Publishers have finished emitting items. *

* *

* Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only * invoke the {@code onError} method of its Subscribers once. *

*
Backpressure:
*
The operator honors backpressure from downstream. All inner {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources * the Iterable of Publishers * @return a Flowable that emits items that are the result of flattening the items emitted by the * Publishers in the Iterable * @see ReactiveX operators documentation: Merge */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeDelayError(Iterable> sources) { return fromIterable(sources).flatMap((Function)Functions.identity(), true); } /** * Flattens an Iterable of Publishers into one Publisher, in a way that allows a Subscriber to receive all * successfully emitted items from each of the source Publishers without being interrupted by an error * notification from one of them, while limiting the number of concurrent subscriptions to these Publishers. *

* This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that * error notification until all of the merged Publishers have finished emitting items. *

* *

* Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only * invoke the {@code onError} method of its Subscribers once. *

*
Backpressure:
*
The operator honors backpressure from downstream. All inner {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources * the Iterable of Publishers * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @param bufferSize * the number of items to prefetch from each inner Publisher * @return a Flowable that emits items that are the result of flattening the items emitted by the * Publishers in the Iterable * @see ReactiveX operators documentation: Merge */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeDelayError(Iterable> sources, int maxConcurrency, int bufferSize) { return fromIterable(sources).flatMap((Function)Functions.identity(), true, maxConcurrency, bufferSize); } /** * Flattens an array of Publishers into one Publisher, in a way that allows a Subscriber to receive all * successfully emitted items from each of the source Publishers without being interrupted by an error * notification from one of them, while limiting the number of concurrent subscriptions to these Publishers. *

* This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that * error notification until all of the merged Publishers have finished emitting items. *

* *

* Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only * invoke the {@code onError} method of its Subscribers once. *

*
Backpressure:
*
The operator honors backpressure from downstream. All source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeArrayDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources * the array of Publishers * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @param bufferSize * the number of items to prefetch from each inner Publisher * @return a Flowable that emits items that are the result of flattening the items emitted by the * Publishers in the Iterable * @see ReactiveX operators documentation: Merge */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeArrayDelayError(int maxConcurrency, int bufferSize, Publisher... sources) { return fromArray(sources).flatMap((Function)Functions.identity(), true, maxConcurrency, bufferSize); } /** * Flattens an Iterable of Publishers into one Publisher, in a way that allows a Subscriber to receive all * successfully emitted items from each of the source Publishers without being interrupted by an error * notification from one of them, while limiting the number of concurrent subscriptions to these Publishers. *

* This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that * error notification until all of the merged Publishers have finished emitting items. *

* *

* Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only * invoke the {@code onError} method of its Subscribers once. *

*
Backpressure:
*
The operator honors backpressure from downstream. All inner {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources * the Iterable of Publishers * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @return a Flowable that emits items that are the result of flattening the items emitted by the * Publishers in the Iterable * @see ReactiveX operators documentation: Merge */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeDelayError(Iterable> sources, int maxConcurrency) { return fromIterable(sources).flatMap((Function)Functions.identity(), true, maxConcurrency); } /** * Flattens a Publisher that emits Publishers into one Publisher, in a way that allows a Subscriber to * receive all successfully emitted items from all of the source Publishers without being interrupted by * an error notification from one of them. *

* This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that * error notification until all of the merged Publishers have finished emitting items. *

* *

* Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only * invoke the {@code onError} method of its Subscribers once. *

*
Backpressure:
*
The operator honors backpressure from downstream. The outer {@code Publisher} is consumed * in unbounded mode (i.e., no backpressure is applied to it). The inner {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources * a Publisher that emits Publishers * @return a Flowable that emits all of the items emitted by the Publishers emitted by the * {@code source} Publisher * @see ReactiveX operators documentation: Merge */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeDelayError(Publisher> sources) { return mergeDelayError(sources, bufferSize()); } /** * Flattens a Publisher that emits Publishers into one Publisher, in a way that allows a Subscriber to * receive all successfully emitted items from all of the source Publishers without being interrupted by * an error notification from one of them, while limiting the * number of concurrent subscriptions to these Publishers. *

* This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that * error notification until all of the merged Publishers have finished emitting items. *

* *

* Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only * invoke the {@code onError} method of its Subscribers once. *

*
Backpressure:
*
The operator honors backpressure from downstream. Both the outer and inner {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources * a Publisher that emits Publishers * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @return a Flowable that emits all of the items emitted by the Publishers emitted by the * {@code source} Publisher * @see ReactiveX operators documentation: Merge * @since 2.0 */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeDelayError(Publisher> sources, int maxConcurrency) { return fromPublisher(sources).flatMap((Function)Functions.identity(), true, maxConcurrency); } /** * Flattens an array of Publishers into one Flowable, in a way that allows a Subscriber to receive all * successfully emitted items from each of the source Publishers without being interrupted by an error * notification from one of them. *

* This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that * error notification until all of the merged Publishers have finished emitting items. *

* *

* Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only * invoke the {@code onError} method of its Subscribers once. *

*
Backpressure:
*
The operator honors backpressure from downstream. Both the outer and inner {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeArrayDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param sources * the Iterable of Publishers * @return a Flowable that emits items that are the result of flattening the items emitted by the * Publishers in the Iterable * @see ReactiveX operators documentation: Merge */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeArrayDelayError(Publisher... sources) { return fromArray(sources).flatMap((Function)Functions.identity(), true, sources.length); } /** * Flattens two Publishers into one Publisher, in a way that allows a Subscriber to receive all * successfully emitted items from each of the source Publishers without being interrupted by an error * notification from one of them. *

* This behaves like {@link #merge(Publisher, Publisher)} except that if any of the merged Publishers * notify of an error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from * propagating that error notification until all of the merged Publishers have finished emitting items. *

* *

* Even if both merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only * invoke the {@code onError} method of its Subscribers once. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param source1 * a Publisher to be merged * @param source2 * a Publisher to be merged * @return a Flowable that emits all of the items that are emitted by the two source Publishers * @see ReactiveX operators documentation: Merge */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeDelayError(Publisher source1, Publisher source2) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); return fromArray(source1, source2).flatMap((Function)Functions.identity(), true, 2); } /** * Flattens three Publishers into one Publisher, in a way that allows a Subscriber to receive all * successfully emitted items from all of the source Publishers without being interrupted by an error * notification from one of them. *

* This behaves like {@link #merge(Publisher, Publisher, Publisher)} except that if any of the merged * Publishers notify of an error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain * from propagating that error notification until all of the merged Publishers have finished emitting * items. *

* *

* Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only * invoke the {@code onError} method of its Subscribers once. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param source1 * a Publisher to be merged * @param source2 * a Publisher to be merged * @param source3 * a Publisher to be merged * @return a Flowable that emits all of the items that are emitted by the source Publishers * @see ReactiveX operators documentation: Merge */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeDelayError(Publisher source1, Publisher source2, Publisher source3) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); return fromArray(source1, source2, source3).flatMap((Function)Functions.identity(), true, 3); } /** * Flattens four Publishers into one Publisher, in a way that allows a Subscriber to receive all * successfully emitted items from all of the source Publishers without being interrupted by an error * notification from one of them. *

* This behaves like {@link #merge(Publisher, Publisher, Publisher, Publisher)} except that if any of * the merged Publishers notify of an error via {@link Subscriber#onError onError}, {@code mergeDelayError} * will refrain from propagating that error notification until all of the merged Publishers have finished * emitting items. *

* *

* Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only * invoke the {@code onError} method of its Subscribers once. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element base type * @param source1 * a Publisher to be merged * @param source2 * a Publisher to be merged * @param source3 * a Publisher to be merged * @param source4 * a Publisher to be merged * @return a Flowable that emits all of the items that are emitted by the source Publishers * @see ReactiveX operators documentation: Merge */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable mergeDelayError( Publisher source1, Publisher source2, Publisher source3, Publisher source4) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); return fromArray(source1, source2, source3, source4).flatMap((Function)Functions.identity(), true, 4); } /** * Returns a Flowable that never sends any items or notifications to a {@link Subscriber}. *

* *

* This Publisher is useful primarily for testing purposes. *

*
Backpressure:
*
This source doesn't produce any elements and effectively ignores downstream backpressure.
*
Scheduler:
*
{@code never} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items (not) emitted by the Publisher * @return a Flowable that never emits any items or sends any notifications to a {@link Subscriber} * @see ReactiveX operators documentation: Never */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) @SuppressWarnings("unchecked") public static Flowable never() { return RxJavaPlugins.onAssembly((Flowable) FlowableNever.INSTANCE); } /** * Returns a Flowable that emits a sequence of Integers within a specified range. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals values on-demand (i.e., when requested).
*
Scheduler:
*
{@code range} does not operate by default on a particular {@link Scheduler}.
*
* * @param start * the value of the first Integer in the sequence * @param count * the number of sequential Integers to generate * @return a Flowable that emits a range of sequential Integers * @throws IllegalArgumentException * if {@code count} is less than zero, or if {@code start} + {@code count} − 1 exceeds * {@code Integer.MAX_VALUE} * @see ReactiveX operators documentation: Range */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable range(int start, int count) { if (count < 0) { throw new IllegalArgumentException("count >= 0 required but it was " + count); } else if (count == 0) { return empty(); } else if (count == 1) { return just(start); } else if ((long)start + (count - 1) > Integer.MAX_VALUE) { throw new IllegalArgumentException("Integer overflow"); } return RxJavaPlugins.onAssembly(new FlowableRange(start, count)); } /** * Returns a Flowable that emits a sequence of Longs within a specified range. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and signals values on-demand (i.e., when requested).
*
Scheduler:
*
{@code rangeLong} does not operate by default on a particular {@link Scheduler}.
*
* * @param start * the value of the first Long in the sequence * @param count * the number of sequential Longs to generate * @return a Flowable that emits a range of sequential Longs * @throws IllegalArgumentException * if {@code count} is less than zero, or if {@code start} + {@code count} − 1 exceeds * {@code Long.MAX_VALUE} * @see ReactiveX operators documentation: Range */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable rangeLong(long start, long count) { if (count < 0) { throw new IllegalArgumentException("count >= 0 required but it was " + count); } if (count == 0) { return empty(); } if (count == 1) { return just(start); } long end = start + (count - 1); if (start > 0 && end < 0) { throw new IllegalArgumentException("Overflow! start + count is bigger than Long.MAX_VALUE"); } return RxJavaPlugins.onAssembly(new FlowableRangeLong(start, count)); } /** * Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the * same by comparing the items emitted by each Publisher pairwise. *

* *

*
Backpressure:
*
This operator honors downstream backpressure and expects both of its sources * to honor backpressure as well. If violated, the operator will emit a MissingBackpressureException.
*
Scheduler:
*
{@code sequenceEqual} does not operate by default on a particular {@link Scheduler}.
*
* * @param source1 * the first Publisher to compare * @param source2 * the second Publisher to compare * @param * the type of items emitted by each Publisher * @return a Flowable that emits a Boolean value that indicates whether the two sequences are the same * @see ReactiveX operators documentation: SequenceEqual */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Single sequenceEqual(Publisher source1, Publisher source2) { return sequenceEqual(source1, source2, ObjectHelper.equalsPredicate(), bufferSize()); } /** * Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the * same by comparing the items emitted by each Publisher pairwise based on the results of a specified * equality function. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator signals a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code sequenceEqual} does not operate by default on a particular {@link Scheduler}.
*
* * @param source1 * the first Publisher to compare * @param source2 * the second Publisher to compare * @param isEqual * a function used to compare items emitted by each Publisher * @param * the type of items emitted by each Publisher * @return a Single that emits a Boolean value that indicates whether the two Publisher sequences * are the same according to the specified function * @see ReactiveX operators documentation: SequenceEqual */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Single sequenceEqual(Publisher source1, Publisher source2, BiPredicate isEqual) { return sequenceEqual(source1, source2, isEqual, bufferSize()); } /** * Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the * same by comparing the items emitted by each Publisher pairwise based on the results of a specified * equality function. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor * backpressure; if violated, the operator signals a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code sequenceEqual} does not operate by default on a particular {@link Scheduler}.
*
* * @param source1 * the first Publisher to compare * @param source2 * the second Publisher to compare * @param isEqual * a function used to compare items emitted by each Publisher * @param bufferSize * the number of items to prefetch from the first and second source Publisher * @param * the type of items emitted by each Publisher * @return a Single that emits a Boolean value that indicates whether the two Publisher sequences * are the same according to the specified function * @see ReactiveX operators documentation: SequenceEqual */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Single sequenceEqual(Publisher source1, Publisher source2, BiPredicate isEqual, int bufferSize) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(isEqual, "isEqual is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableSequenceEqualSingle(source1, source2, isEqual, bufferSize)); } /** * Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the * same by comparing the items emitted by each Publisher pairwise. *

* *

*
Backpressure:
*
This operator honors downstream backpressure and expects both of its sources * to honor backpressure as well. If violated, the operator will emit a MissingBackpressureException.
*
Scheduler:
*
{@code sequenceEqual} does not operate by default on a particular {@link Scheduler}.
*
* * @param source1 * the first Publisher to compare * @param source2 * the second Publisher to compare * @param bufferSize * the number of items to prefetch from the first and second source Publisher * @param * the type of items emitted by each Publisher * @return a Single that emits a Boolean value that indicates whether the two sequences are the same * @see ReactiveX operators documentation: SequenceEqual */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Single sequenceEqual(Publisher source1, Publisher source2, int bufferSize) { return sequenceEqual(source1, source2, ObjectHelper.equalsPredicate(), bufferSize); } /** * Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the * most recently emitted of those Publishers. *

* *

* {@code switchOnNext} subscribes to a Publisher that emits Publishers. Each time it observes one of * these emitted Publishers, the Publisher returned by {@code switchOnNext} begins emitting the items * emitted by that Publisher. When a new Publisher is emitted, {@code switchOnNext} stops emitting items * from the earlier-emitted Publisher and begins emitting items from the new one. *

* The resulting Publisher completes if both the outer Publisher and the last inner Publisher, if any, complete. * If the outer Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence. *

*
Backpressure:
*
The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an * unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor * backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException} * but the violation may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code switchOnNext} does not operate by default on a particular {@link Scheduler}.
*
* * @param the item type * @param sources * the source Publisher that emits Publishers * @param bufferSize * the number of items to prefetch from the inner Publishers * @return a Flowable that emits the items emitted by the Publisher most recently emitted by the source * Publisher * @see ReactiveX operators documentation: Switch */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable switchOnNext(Publisher> sources, int bufferSize) { return fromPublisher(sources).switchMap((Function)Functions.identity(), bufferSize); } /** * Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the * most recently emitted of those Publishers. *

* *

* {@code switchOnNext} subscribes to a Publisher that emits Publishers. Each time it observes one of * these emitted Publishers, the Publisher returned by {@code switchOnNext} begins emitting the items * emitted by that Publisher. When a new Publisher is emitted, {@code switchOnNext} stops emitting items * from the earlier-emitted Publisher and begins emitting items from the new one. *

* The resulting Publisher completes if both the outer Publisher and the last inner Publisher, if any, complete. * If the outer Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence. *

*
Backpressure:
*
The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an * unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor * backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException} * but the violation may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code switchOnNext} does not operate by default on a particular {@link Scheduler}.
*
* * @param the item type * @param sources * the source Publisher that emits Publishers * @return a Flowable that emits the items emitted by the Publisher most recently emitted by the source * Publisher * @see ReactiveX operators documentation: Switch */ @SuppressWarnings({ "unchecked", "rawtypes" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable switchOnNext(Publisher> sources) { return fromPublisher(sources).switchMap((Function)Functions.identity()); } /** * Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the * most recently emitted of those Publishers and delays any exception until all Publishers terminate. *

* *

* {@code switchOnNext} subscribes to a Publisher that emits Publishers. Each time it observes one of * these emitted Publishers, the Publisher returned by {@code switchOnNext} begins emitting the items * emitted by that Publisher. When a new Publisher is emitted, {@code switchOnNext} stops emitting items * from the earlier-emitted Publisher and begins emitting items from the new one. *

* The resulting Publisher completes if both the main Publisher and the last inner Publisher, if any, complete. * If the main Publisher signals an onError, the termination of the last inner Publisher will emit that error as is * or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled. *

*
Backpressure:
*
The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an * unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor * backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException} * but the violation may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code switchOnNextDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the item type * @param sources * the source Publisher that emits Publishers * @return a Flowable that emits the items emitted by the Publisher most recently emitted by the source * Publisher * @see ReactiveX operators documentation: Switch * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable switchOnNextDelayError(Publisher> sources) { return switchOnNextDelayError(sources, bufferSize()); } /** * Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the * most recently emitted of those Publishers and delays any exception until all Publishers terminate. *

* *

* {@code switchOnNext} subscribes to a Publisher that emits Publishers. Each time it observes one of * these emitted Publishers, the Publisher returned by {@code switchOnNext} begins emitting the items * emitted by that Publisher. When a new Publisher is emitted, {@code switchOnNext} stops emitting items * from the earlier-emitted Publisher and begins emitting items from the new one. *

* The resulting Publisher completes if both the main Publisher and the last inner Publisher, if any, complete. * If the main Publisher signals an onError, the termination of the last inner Publisher will emit that error as is * or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled. *

*
Backpressure:
*
The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an * unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor * backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException} * but the violation may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code switchOnNextDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the item type * @param sources * the source Publisher that emits Publishers * @param prefetch * the number of items to prefetch from the inner Publishers * @return a Flowable that emits the items emitted by the Publisher most recently emitted by the source * Publisher * @see ReactiveX operators documentation: Switch * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable switchOnNextDelayError(Publisher> sources, int prefetch) { return fromPublisher(sources).switchMapDelayError(Functions.>identity(), prefetch); } /** * Returns a Flowable that emits {@code 0L} after a specified delay, and then completes. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time. If the downstream needs a slower rate * it should slow the timer or use something like {@link #onBackpressureDrop}.
*
Scheduler:
*
{@code timer} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param delay * the initial delay before emitting a single {@code 0L} * @param unit * time units to use for {@code delay} * @return a Flowable that emits {@code 0L} after a specified delay, and then completes * @see ReactiveX operators documentation: Timer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public static Flowable timer(long delay, TimeUnit unit) { return timer(delay, unit, Schedulers.computation()); } /** * Returns a Flowable that emits {@code 0L} after a specified delay, on a specified Scheduler, and then * completes. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time. If the downstream needs a slower rate * it should slow the timer or use something like {@link #onBackpressureDrop}.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param delay * the initial delay before emitting a single 0L * @param unit * time units to use for {@code delay} * @param scheduler * the {@link Scheduler} to use for scheduling the item * @return a Flowable that emits {@code 0L} after a specified delay, on a specified Scheduler, and then * completes * @see ReactiveX operators documentation: Timer */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public static Flowable timer(long delay, TimeUnit unit, Scheduler scheduler) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableTimer(Math.max(0L, delay), unit, scheduler)); } /** * Create a Flowable by wrapping a Publisher which has to be implemented according * to the Reactive Streams specification by handling backpressure and * cancellation correctly; no safeguards are provided by the Flowable itself. *
*
Backpressure:
*
This operator is a pass-through for backpressure and the behavior is determined by the * provided Publisher implementation.
*
Scheduler:
*
{@code unsafeCreate} by default doesn't operate on any particular {@link Scheduler}.
*
* @param the value type emitted * @param onSubscribe the Publisher instance to wrap * @return the new Flowable instance * @throws IllegalArgumentException if {@code onSubscribe} is a subclass of {@code Flowable}; such * instances don't need conversion and is possibly a port remnant from 1.x or one should use {@link #hide()} * instead. */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.NONE) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable unsafeCreate(Publisher onSubscribe) { ObjectHelper.requireNonNull(onSubscribe, "onSubscribe is null"); if (onSubscribe instanceof Flowable) { throw new IllegalArgumentException("unsafeCreate(Flowable) should be upgraded"); } return RxJavaPlugins.onAssembly(new FlowableFromPublisher(onSubscribe)); } /** * Constructs a Publisher that creates a dependent resource object which is disposed of on cancellation. *

* *

*
Backpressure:
*
The operator is a pass-through for backpressure and otherwise depends on the * backpressure support of the Publisher returned by the {@code resourceFactory}.
*
Scheduler:
*
{@code using} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the generated Publisher * @param the type of the resource associated with the output sequence * @param resourceSupplier * the factory function to create a resource object that depends on the Publisher * @param sourceSupplier * the factory function to create a Publisher * @param resourceDisposer * the function that will dispose of the resource * @return the Publisher whose lifetime controls the lifetime of the dependent resource object * @see ReactiveX operators documentation: Using */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable using(Callable resourceSupplier, Function> sourceSupplier, Consumer resourceDisposer) { return using(resourceSupplier, sourceSupplier, resourceDisposer, true); } /** * Constructs a Publisher that creates a dependent resource object which is disposed of just before * termination if you have set {@code disposeEagerly} to {@code true} and cancellation does not occur * before termination. Otherwise, resource disposal will occur on cancellation. Eager disposal is * particularly appropriate for a synchronous Publisher that reuses resources. {@code disposeAction} will * only be called once per subscription. *

* *

*
Backpressure:
*
The operator is a pass-through for backpressure and otherwise depends on the * backpressure support of the Publisher returned by the {@code resourceFactory}.
*
Scheduler:
*
{@code using} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the generated Publisher * @param the type of the resource associated with the output sequence * @param resourceSupplier * the factory function to create a resource object that depends on the Publisher * @param sourceSupplier * the factory function to create a Publisher * @param resourceDisposer * the function that will dispose of the resource * @param eager * if {@code true} then disposal will happen either on cancellation or just before emission of * a terminal event ({@code onComplete} or {@code onError}). * @return the Publisher whose lifetime controls the lifetime of the dependent resource object * @see ReactiveX operators documentation: Using * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable using(Callable resourceSupplier, Function> sourceSupplier, Consumer resourceDisposer, boolean eager) { ObjectHelper.requireNonNull(resourceSupplier, "resourceSupplier is null"); ObjectHelper.requireNonNull(sourceSupplier, "sourceSupplier is null"); ObjectHelper.requireNonNull(resourceDisposer, "resourceDisposer is null"); return RxJavaPlugins.onAssembly(new FlowableUsing(resourceSupplier, sourceSupplier, resourceDisposer, eager)); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * items emitted, in sequence, by an Iterable of other Publishers. *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by each of the source Publishers; * the second item emitted by the new Publisher will be the result of the function applied to the second * item emitted by each of those Publishers; and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@code onNext} as many times as * the number of {@code onNext} invocations of the source Publisher that emits the fewest items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(Arrays.asList(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)), (a) -> a)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *

* *

*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common value type * @param the zipped result type * @param sources * an Iterable of source Publishers * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results in * an item that will be emitted by the resulting Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip(Iterable> sources, Function zipper) { ObjectHelper.requireNonNull(zipper, "zipper is null"); ObjectHelper.requireNonNull(sources, "sources is null"); return RxJavaPlugins.onAssembly(new FlowableZip(null, sources, zipper, bufferSize(), false)); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * n items emitted, in sequence, by the n Publishers emitted by a specified Publisher. *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by each of the Publishers emitted * by the source Publisher; the second item emitted by the new Publisher will be the result of the * function applied to the second item emitted by each of those Publishers; and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@code onNext} as many times as * the number of {@code onNext} invocations of the source Publisher that emits the fewest items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while cancel the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(just(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)), (a) -> a)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *

* *

*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the inner Publishers * @param the zipped result type * @param sources * a Publisher of source Publishers * @param zipper * a function that, when applied to an item emitted by each of the Publishers emitted by * {@code ws}, results in an item that will be emitted by the resulting Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings({ "rawtypes", "unchecked", "cast" }) @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip(Publisher> sources, final Function zipper) { ObjectHelper.requireNonNull(zipper, "zipper is null"); return fromPublisher(sources).toList().flatMapPublisher((Function)FlowableInternalHelper.zipIterable(zipper)); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * two items emitted, in sequence, by two other Publishers. *

* *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by {@code o1} and the first item * emitted by {@code o2}; the second item emitted by the new Publisher will be the result of the function * applied to the second item emitted by {@code o1} and the second item emitted by {@code o2}; and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@link Subscriber#onNext onNext} * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest * items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), (a, b) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *
*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the first source * @param the value type of the second source * @param the zipped result type * @param source1 * the first source Publisher * @param source2 * a second source Publisher * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results * in an item that will be emitted by the resulting Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip( Publisher source1, Publisher source2, BiFunction zipper) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * two items emitted, in sequence, by two other Publishers. *

* *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by {@code o1} and the first item * emitted by {@code o2}; the second item emitted by the new Publisher will be the result of the function * applied to the second item emitted by {@code o1} and the second item emitted by {@code o2}; and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@link Subscriber#onNext onNext} * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest * items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), (a, b) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *
*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the first source * @param the value type of the second source * @param the zipped result type * @param source1 * the first source Publisher * @param source2 * a second source Publisher * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results * in an item that will be emitted by the resulting Publisher * @param delayError delay errors from any of the source Publishers till the other terminates * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip( Publisher source1, Publisher source2, BiFunction zipper, boolean delayError) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); return zipArray(Functions.toFunction(zipper), delayError, bufferSize(), source1, source2); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * two items emitted, in sequence, by two other Publishers. *

* *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by {@code o1} and the first item * emitted by {@code o2}; the second item emitted by the new Publisher will be the result of the function * applied to the second item emitted by {@code o1} and the second item emitted by {@code o2}; and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@link Subscriber#onNext onNext} * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest * items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), (a, b) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *
*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the first source * @param the value type of the second source * @param the zipped result type * @param source1 * the first source Publisher * @param source2 * a second source Publisher * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results * in an item that will be emitted by the resulting Publisher * @param delayError delay errors from any of the source Publishers till the other terminates * @param bufferSize the number of elements to prefetch from each source Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip( Publisher source1, Publisher source2, BiFunction zipper, boolean delayError, int bufferSize) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); return zipArray(Functions.toFunction(zipper), delayError, bufferSize, source1, source2); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * three items emitted, in sequence, by three other Publishers. *

* *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by {@code o1}, the first item * emitted by {@code o2}, and the first item emitted by {@code o3}; the second item emitted by the new * Publisher will be the result of the function applied to the second item emitted by {@code o1}, the * second item emitted by {@code o2}, and the second item emitted by {@code o3}; and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@link Subscriber#onNext onNext} * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest * items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *
*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the first source * @param the value type of the second source * @param the value type of the third source * @param the zipped result type * @param source1 * the first source Publisher * @param source2 * a second source Publisher * @param source3 * a third source Publisher * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results in * an item that will be emitted by the resulting Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip( Publisher source1, Publisher source2, Publisher source3, Function3 zipper) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * four items emitted, in sequence, by four other Publishers. *

* *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by {@code o1}, the first item * emitted by {@code o2}, the first item emitted by {@code o3}, and the first item emitted by {@code 04}; * the second item emitted by the new Publisher will be the result of the function applied to the second * item emitted by each of those Publishers; and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@link Subscriber#onNext onNext} * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest * items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *
*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the first source * @param the value type of the second source * @param the value type of the third source * @param the value type of the fourth source * @param the zipped result type * @param source1 * the first source Publisher * @param source2 * a second source Publisher * @param source3 * a third source Publisher * @param source4 * a fourth source Publisher * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results in * an item that will be emitted by the resulting Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Function4 zipper) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * five items emitted, in sequence, by five other Publishers. *

* *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by {@code o1}, the first item * emitted by {@code o2}, the first item emitted by {@code o3}, the first item emitted by {@code o4}, and * the first item emitted by {@code o5}; the second item emitted by the new Publisher will be the result of * the function applied to the second item emitted by each of those Publishers; and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@link Subscriber#onNext onNext} * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest * items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *
*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the first source * @param the value type of the second source * @param the value type of the third source * @param the value type of the fourth source * @param the value type of the fifth source * @param the zipped result type * @param source1 * the first source Publisher * @param source2 * a second source Publisher * @param source3 * a third source Publisher * @param source4 * a fourth source Publisher * @param source5 * a fifth source Publisher * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results in * an item that will be emitted by the resulting Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Publisher source5, Function5 zipper) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); ObjectHelper.requireNonNull(source5, "source5 is null"); return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4, source5); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * six items emitted, in sequence, by six other Publishers. *

* *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by each source Publisher, the * second item emitted by the new Publisher will be the result of the function applied to the second item * emitted by each of those Publishers, and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@link Subscriber#onNext onNext} * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest * items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *
*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the first source * @param the value type of the second source * @param the value type of the third source * @param the value type of the fourth source * @param the value type of the fifth source * @param the value type of the sixth source * @param the zipped result type * @param source1 * the first source Publisher * @param source2 * a second source Publisher * @param source3 * a third source Publisher * @param source4 * a fourth source Publisher * @param source5 * a fifth source Publisher * @param source6 * a sixth source Publisher * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results in * an item that will be emitted by the resulting Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Publisher source5, Publisher source6, Function6 zipper) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); ObjectHelper.requireNonNull(source5, "source5 is null"); ObjectHelper.requireNonNull(source6, "source6 is null"); return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4, source5, source6); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * seven items emitted, in sequence, by seven other Publishers. *

* *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by each source Publisher, the * second item emitted by the new Publisher will be the result of the function applied to the second item * emitted by each of those Publishers, and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@link Subscriber#onNext onNext} * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest * items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f, g) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *
*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the first source * @param the value type of the second source * @param the value type of the third source * @param the value type of the fourth source * @param the value type of the fifth source * @param the value type of the sixth source * @param the value type of the seventh source * @param the zipped result type * @param source1 * the first source Publisher * @param source2 * a second source Publisher * @param source3 * a third source Publisher * @param source4 * a fourth source Publisher * @param source5 * a fifth source Publisher * @param source6 * a sixth source Publisher * @param source7 * a seventh source Publisher * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results in * an item that will be emitted by the resulting Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Publisher source5, Publisher source6, Publisher source7, Function7 zipper) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); ObjectHelper.requireNonNull(source5, "source5 is null"); ObjectHelper.requireNonNull(source6, "source6 is null"); ObjectHelper.requireNonNull(source7, "source7 is null"); return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4, source5, source6, source7); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * eight items emitted, in sequence, by eight other Publishers. *

* *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by each source Publisher, the * second item emitted by the new Publisher will be the result of the function applied to the second item * emitted by each of those Publishers, and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@link Subscriber#onNext onNext} * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest * items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f, g, h) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *
*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the first source * @param the value type of the second source * @param the value type of the third source * @param the value type of the fourth source * @param the value type of the fifth source * @param the value type of the sixth source * @param the value type of the seventh source * @param the value type of the eighth source * @param the zipped result type * @param source1 * the first source Publisher * @param source2 * a second source Publisher * @param source3 * a third source Publisher * @param source4 * a fourth source Publisher * @param source5 * a fifth source Publisher * @param source6 * a sixth source Publisher * @param source7 * a seventh source Publisher * @param source8 * an eighth source Publisher * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results in * an item that will be emitted by the resulting Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Publisher source5, Publisher source6, Publisher source7, Publisher source8, Function8 zipper) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); ObjectHelper.requireNonNull(source5, "source5 is null"); ObjectHelper.requireNonNull(source6, "source6 is null"); ObjectHelper.requireNonNull(source7, "source7 is null"); ObjectHelper.requireNonNull(source8, "source8 is null"); return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4, source5, source6, source7, source8); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * nine items emitted, in sequence, by nine other Publishers. *

* *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by each source Publisher, the * second item emitted by the new Publisher will be the result of the function applied to the second item * emitted by each of those Publishers, and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@link Subscriber#onNext onNext} * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest * items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f, g, h, i) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *
*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zip} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the first source * @param the value type of the second source * @param the value type of the third source * @param the value type of the fourth source * @param the value type of the fifth source * @param the value type of the sixth source * @param the value type of the seventh source * @param the value type of the eighth source * @param the value type of the ninth source * @param the zipped result type * @param source1 * the first source Publisher * @param source2 * a second source Publisher * @param source3 * a third source Publisher * @param source4 * a fourth source Publisher * @param source5 * a fifth source Publisher * @param source6 * a sixth source Publisher * @param source7 * a seventh source Publisher * @param source8 * an eighth source Publisher * @param source9 * a ninth source Publisher * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results in * an item that will be emitted by the resulting Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zip( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Publisher source5, Publisher source6, Publisher source7, Publisher source8, Publisher source9, Function9 zipper) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); ObjectHelper.requireNonNull(source5, "source5 is null"); ObjectHelper.requireNonNull(source6, "source6 is null"); ObjectHelper.requireNonNull(source7, "source7 is null"); ObjectHelper.requireNonNull(source8, "source8 is null"); ObjectHelper.requireNonNull(source9, "source9 is null"); return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4, source5, source6, source7, source8, source9); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * items emitted, in sequence, by an array of other Publishers. *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by each of the source Publishers; * the second item emitted by the new Publisher will be the result of the function applied to the second * item emitted by each of those Publishers; and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@code onNext} as many times as * the number of {@code onNext} invocations of the source Publisher that emits the fewest items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(new Publisher[]{range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)}, (a) ->
     * a)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *

* *

*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zipArray} does not operate by default on a particular {@link Scheduler}.
*
* * @param the common element type * @param the result type * @param sources * an array of source Publishers * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results in * an item that will be emitted by the resulting Publisher * @param delayError * delay errors signaled by any of the source Publisher until all Publishers terminate * @param bufferSize * the number of elements to prefetch from each source Publisher * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zipArray(Function zipper, boolean delayError, int bufferSize, Publisher... sources) { if (sources.length == 0) { return empty(); } ObjectHelper.requireNonNull(zipper, "zipper is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableZip(sources, null, zipper, bufferSize, delayError)); } /** * Returns a Flowable that emits the results of a specified combiner function applied to combinations of * items emitted, in sequence, by an Iterable of other Publishers. *

* {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher * will be the result of the function applied to the first item emitted by each of the source Publishers; * the second item emitted by the new Publisher will be the result of the function applied to the second * item emitted by each of those Publishers; and so forth. *

* The resulting {@code Publisher} returned from {@code zip} will invoke {@code onNext} as many times as * the number of {@code onNext} invocations of the source Publisher that emits the fewest items. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

zip(Arrays.asList(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)), (a) -> a)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *

* *

*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zipIterable} does not operate by default on a particular {@link Scheduler}.
*
* * * @param sources * an Iterable of source Publishers * @param zipper * a function that, when applied to an item emitted by each of the source Publishers, results in * an item that will be emitted by the resulting Publisher * @param delayError * delay errors signaled by any of the source Publisher until all Publishers terminate * @param bufferSize * the number of elements to prefetch from each source Publisher * @param the common source value type * @param the zipped result type * @return a Flowable that emits the zipped results * @see ReactiveX operators documentation: Zip */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public static Flowable zipIterable(Iterable> sources, Function zipper, boolean delayError, int bufferSize) { ObjectHelper.requireNonNull(zipper, "zipper is null"); ObjectHelper.requireNonNull(sources, "sources is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableZip(null, sources, zipper, bufferSize, delayError)); } // *************************************************************************************************** // Instance operators // *************************************************************************************************** /** * Returns a Single that emits a Boolean that indicates whether all of the items emitted by the source * Publisher satisfy a condition. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code all} does not operate by default on a particular {@link Scheduler}.
*
* * @param predicate * a function that evaluates an item and returns a Boolean * @return a Single that emits {@code true} if all items emitted by the source Publisher satisfy the * predicate; otherwise, {@code false} * @see ReactiveX operators documentation: All */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single all(Predicate predicate) { ObjectHelper.requireNonNull(predicate, "predicate is null"); return RxJavaPlugins.onAssembly(new FlowableAllSingle(this, predicate)); } /** * Mirrors the Publisher (current or provided) that first either emits an item or sends a termination * notification. *

* *

*
Backpressure:
*
The operator itself doesn't interfere with backpressure which is determined by the winning * {@code Publisher}'s backpressure behavior.
*
Scheduler:
*
{@code ambWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param other * a Publisher competing to react first. A subscription to this provided Publisher will occur after subscribing * to the current Publisher. * @return a Flowable that emits the same sequence as whichever of the source Publishers first * emitted an item or sent a termination notification * @see ReactiveX operators documentation: Amb */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable ambWith(Publisher other) { ObjectHelper.requireNonNull(other, "other is null"); return ambArray(this, other); } /** * Returns a Single that emits {@code true} if any item emitted by the source Publisher satisfies a * specified condition, otherwise {@code false}. Note: this always emits {@code false} if the * source Publisher is empty. *

* *

* In Rx.Net this is the {@code any} operator but we renamed it in RxJava to better match Java naming * idioms. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code any} does not operate by default on a particular {@link Scheduler}.
*
* * @param predicate * the condition to test items emitted by the source Publisher * @return a Single that emits a Boolean that indicates whether any item emitted by the source * Publisher satisfies the {@code predicate} * @see ReactiveX operators documentation: Contains */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single any(Predicate predicate) { ObjectHelper.requireNonNull(predicate, "predicate is null"); return RxJavaPlugins.onAssembly(new FlowableAnySingle(this, predicate)); } /** * Calls the specified converter function during assembly time and returns its resulting value. *

* This allows fluent conversion to any other type. *

*
Backpressure:
*
The backpressure behavior depends on what happens in the {@code converter} function.
*
Scheduler:
*
{@code as} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.7 - experimental * @param the resulting object type * @param converter the function that receives the current Flowable instance and returns a value * @return the converted value * @throws NullPointerException if converter is null * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final R as(@NonNull FlowableConverter converter) { return ObjectHelper.requireNonNull(converter, "converter is null").apply(this); } /** * Returns the first item emitted by this {@code Flowable}, or throws * {@code NoSuchElementException} if it emits no items. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingFirst} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If the source signals an error, the operator wraps a checked {@link Exception} * into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and * {@link Error}s are rethrown as they are.
*
* * @return the first item emitted by this {@code Flowable} * @throws NoSuchElementException * if this {@code Flowable} emits no items * @see ReactiveX documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final T blockingFirst() { BlockingFirstSubscriber s = new BlockingFirstSubscriber(); subscribe(s); T v = s.blockingGet(); if (v != null) { return v; } throw new NoSuchElementException(); } /** * Returns the first item emitted by this {@code Flowable}, or a default value if it emits no * items. *
*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingFirst} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If the source signals an error, the operator wraps a checked {@link Exception} * into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and * {@link Error}s are rethrown as they are.
*
* * @param defaultItem * a default value to return if this {@code Flowable} emits no items * @return the first item emitted by this {@code Flowable}, or the default value if it emits no * items * @see ReactiveX documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final T blockingFirst(T defaultItem) { BlockingFirstSubscriber s = new BlockingFirstSubscriber(); subscribe(s); T v = s.blockingGet(); return v != null ? v : defaultItem; } /** * Consumes the upstream {@code Flowable} in a blocking fashion and invokes the given * {@code Consumer} with each upstream item on the current thread until the * upstream terminates. *

* *

* Note: the method will only return if the upstream terminates or the current * thread is interrupted. *

* This method executes the {@code Consumer} on the current thread while * {@link #subscribe(Consumer)} executes the consumer on the original caller thread of the * sequence. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingForEach} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If the source signals an error, the operator wraps a checked {@link Exception} * into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and * {@link Error}s are rethrown as they are.
*
* * @param onNext * the {@link Consumer} to invoke for each item emitted by the {@code Flowable} * @throws RuntimeException * if an error occurs * @see ReactiveX documentation: Subscribe * @see #subscribe(Consumer) */ @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final void blockingForEach(Consumer onNext) { Iterator it = blockingIterable().iterator(); while (it.hasNext()) { try { onNext.accept(it.next()); } catch (Throwable e) { Exceptions.throwIfFatal(e); ((Disposable)it).dispose(); throw ExceptionHelper.wrapOrThrow(e); } } } /** * Converts this {@code Flowable} into an {@link Iterable}. *

* *

*
Backpressure:
*
The operator expects the upstream to honor backpressure otherwise the returned * Iterable's iterator will throw a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code blockingIterable} does not operate by default on a particular {@link Scheduler}.
*
* * @return an {@link Iterable} version of this {@code Flowable} * @see ReactiveX documentation: To */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Iterable blockingIterable() { return blockingIterable(bufferSize()); } /** * Converts this {@code Flowable} into an {@link Iterable}. *

* *

*
Backpressure:
*
The operator expects the upstream to honor backpressure otherwise the returned * Iterable's iterator will throw a {@code MissingBackpressureException}. *
*
Scheduler:
*
{@code blockingIterable} does not operate by default on a particular {@link Scheduler}.
*
* * @param bufferSize the number of items to prefetch from the current Flowable * @return an {@link Iterable} version of this {@code Flowable} * @see ReactiveX documentation: To */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Iterable blockingIterable(int bufferSize) { ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return new BlockingFlowableIterable(this, bufferSize); } /** * Returns the last item emitted by this {@code Flowable}, or throws * {@code NoSuchElementException} if this {@code Flowable} emits no items. *

* *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingLast} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If the source signals an error, the operator wraps a checked {@link Exception} * into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and * {@link Error}s are rethrown as they are.
*
* * @return the last item emitted by this {@code Flowable} * @throws NoSuchElementException * if this {@code Flowable} emits no items * @see ReactiveX documentation: Last */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final T blockingLast() { BlockingLastSubscriber s = new BlockingLastSubscriber(); subscribe(s); T v = s.blockingGet(); if (v != null) { return v; } throw new NoSuchElementException(); } /** * Returns the last item emitted by this {@code Flowable}, or a default value if it emits no * items. *

* *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingLast} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If the source signals an error, the operator wraps a checked {@link Exception} * into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and * {@link Error}s are rethrown as they are.
*
* * @param defaultItem * a default value to return if this {@code Flowable} emits no items * @return the last item emitted by the {@code Flowable}, or the default value if it emits no * items * @see ReactiveX documentation: Last */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final T blockingLast(T defaultItem) { BlockingLastSubscriber s = new BlockingLastSubscriber(); subscribe(s); T v = s.blockingGet(); return v != null ? v : defaultItem; } /** * Returns an {@link Iterable} that returns the latest item emitted by this {@code Flowable}, * waiting if necessary for one to become available. *

* If this {@code Flowable} produces items faster than {@code Iterator.next} takes them, * {@code onNext} events might be skipped, but {@code onError} or {@code onComplete} events are not. *

* Note also that an {@code onNext} directly followed by {@code onComplete} might hide the {@code onNext} * event. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @return an Iterable that always returns the latest item emitted by this {@code Flowable} * @see ReactiveX documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Iterable blockingLatest() { return new BlockingFlowableLatest(this); } /** * Returns an {@link Iterable} that always returns the item most recently emitted by this * {@code Flowable}. *

* *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingMostRecent} does not operate by default on a particular {@link Scheduler}.
*
* * @param initialItem * the initial item that the {@link Iterable} sequence will yield if this * {@code Flowable} has not yet emitted an item * @return an {@link Iterable} that on each iteration returns the item that this {@code Flowable} * has most recently emitted * @see ReactiveX documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Iterable blockingMostRecent(T initialItem) { return new BlockingFlowableMostRecent(this, initialItem); } /** * Returns an {@link Iterable} that blocks until this {@code Flowable} emits another item, then * returns that item. *

* *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingNext} does not operate by default on a particular {@link Scheduler}.
*
* * @return an {@link Iterable} that blocks upon each iteration until this {@code Flowable} emits * a new item, whereupon the Iterable returns that item * @see ReactiveX documentation: TakeLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Iterable blockingNext() { return new BlockingFlowableNext(this); } /** * If this {@code Flowable} completes after emitting a single item, return that item, otherwise * throw a {@code NoSuchElementException}. *

* *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingSingle} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If the source signals an error, the operator wraps a checked {@link Exception} * into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and * {@link Error}s are rethrown as they are.
*
* * @return the single item emitted by this {@code Flowable} * @see ReactiveX documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final T blockingSingle() { return singleOrError().blockingGet(); } /** * If this {@code Flowable} completes after emitting a single item, return that item; if it emits * more than one item, throw an {@code IllegalArgumentException}; if it emits no items, return a default * value. *

* *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingSingle} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If the source signals an error, the operator wraps a checked {@link Exception} * into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and * {@link Error}s are rethrown as they are.
*
* * @param defaultItem * a default value to return if this {@code Flowable} emits no items * @return the single item emitted by this {@code Flowable}, or the default value if it emits no * items * @see ReactiveX documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final T blockingSingle(T defaultItem) { return single(defaultItem).blockingGet(); } /** * Returns a {@link Future} representing the only value emitted by this {@code Flowable}. *

* *

* If the {@link Flowable} emits more than one item, {@link java.util.concurrent.Future} will receive an * {@link java.lang.IndexOutOfBoundsException}. If the {@link Flowable} is empty, {@link java.util.concurrent.Future} * will receive a {@link java.util.NoSuchElementException}. The {@code Flowable} source has to terminate in order * for the returned {@code Future} to terminate as well. *

* If the {@code Flowable} may emit more than one item, use {@code Flowable.toList().toFuture()}. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code toFuture} does not operate by default on a particular {@link Scheduler}.
*
* * @return a {@link Future} that expects a single item to be emitted by this {@code Flowable} * @see ReactiveX documentation: To */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Future toFuture() { return subscribeWith(new FutureSubscriber()); } /** * Runs the source Flowable to a terminal event, ignoring any values and rethrowing any exception. *

* Note that calling this method will block the caller thread until the upstream terminates * normally or with an error. Therefore, calling this method from special threads such as the * Android Main Thread or the Swing Event Dispatch Thread is not recommended. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.
*
* @since 2.0 * @see #blockingSubscribe(Consumer) * @see #blockingSubscribe(Consumer, Consumer) * @see #blockingSubscribe(Consumer, Consumer, Action) */ @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final void blockingSubscribe() { FlowableBlockingSubscribe.subscribe(this); } /** * Subscribes to the source and calls the given callbacks on the current thread. *

* If the Flowable emits an error, it is wrapped into an * {@link io.reactivex.exceptions.OnErrorNotImplementedException OnErrorNotImplementedException} * and routed to the RxJavaPlugins.onError handler. * Using the overloads {@link #blockingSubscribe(Consumer, Consumer)} * or {@link #blockingSubscribe(Consumer, Consumer, Action)} instead is recommended. *

* Note that calling this method will block the caller thread until the upstream terminates * normally or with an error. Therefore, calling this method from special threads such as the * Android Main Thread or the Swing Event Dispatch Thread is not recommended. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.
*
* @param onNext the callback action for each source value * @since 2.0 * @see #blockingSubscribe(Consumer, Consumer) * @see #blockingSubscribe(Consumer, Consumer, Action) */ @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final void blockingSubscribe(Consumer onNext) { FlowableBlockingSubscribe.subscribe(this, onNext, Functions.ON_ERROR_MISSING, Functions.EMPTY_ACTION); } /** * Subscribes to the source and calls the given callbacks on the current thread. *

* If the Flowable emits an error, it is wrapped into an * {@link io.reactivex.exceptions.OnErrorNotImplementedException OnErrorNotImplementedException} * and routed to the RxJavaPlugins.onError handler. * Using the overloads {@link #blockingSubscribe(Consumer, Consumer)} * or {@link #blockingSubscribe(Consumer, Consumer, Action)} instead is recommended. *

* Note that calling this method will block the caller thread until the upstream terminates * normally or with an error. Therefore, calling this method from special threads such as the * Android Main Thread or the Swing Event Dispatch Thread is not recommended. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an bounded manner (up to bufferSize * outstanding request amount for items).
*
Scheduler:
*
{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.15 - experimental * @param onNext the callback action for each source value * @param bufferSize the size of the buffer * @see #blockingSubscribe(Consumer, Consumer) * @see #blockingSubscribe(Consumer, Consumer, Action) * @since 2.2 */ @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final void blockingSubscribe(Consumer onNext, int bufferSize) { FlowableBlockingSubscribe.subscribe(this, onNext, Functions.ON_ERROR_MISSING, Functions.EMPTY_ACTION, bufferSize); } /** * Subscribes to the source and calls the given callbacks on the current thread. *

* Note that calling this method will block the caller thread until the upstream terminates * normally or with an error. Therefore, calling this method from special threads such as the * Android Main Thread or the Swing Event Dispatch Thread is not recommended. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.
*
* @param onNext the callback action for each source value * @param onError the callback action for an error event * @since 2.0 * @see #blockingSubscribe(Consumer, Consumer, Action) */ @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final void blockingSubscribe(Consumer onNext, Consumer onError) { FlowableBlockingSubscribe.subscribe(this, onNext, onError, Functions.EMPTY_ACTION); } /** * Subscribes to the source and calls the given callbacks on the current thread. *

* Note that calling this method will block the caller thread until the upstream terminates * normally or with an error. Therefore, calling this method from special threads such as the * Android Main Thread or the Swing Event Dispatch Thread is not recommended. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an bounded manner (up to bufferSize * outstanding request amount for items).
*
Scheduler:
*
{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.15 - experimental * @param onNext the callback action for each source value * @param onError the callback action for an error event * @param bufferSize the size of the buffer * @since 2.2 * @see #blockingSubscribe(Consumer, Consumer, Action) */ @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final void blockingSubscribe(Consumer onNext, Consumer onError, int bufferSize) { FlowableBlockingSubscribe.subscribe(this, onNext, onError, Functions.EMPTY_ACTION, bufferSize); } /** * Subscribes to the source and calls the given callbacks on the current thread. *

* Note that calling this method will block the caller thread until the upstream terminates * normally or with an error. Therefore, calling this method from special threads such as the * Android Main Thread or the Swing Event Dispatch Thread is not recommended. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.
*
* @param onNext the callback action for each source value * @param onError the callback action for an error event * @param onComplete the callback action for the completion event. * @since 2.0 */ @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final void blockingSubscribe(Consumer onNext, Consumer onError, Action onComplete) { FlowableBlockingSubscribe.subscribe(this, onNext, onError, onComplete); } /** * Subscribes to the source and calls the given callbacks on the current thread. *

* Note that calling this method will block the caller thread until the upstream terminates * normally or with an error. Therefore, calling this method from special threads such as the * Android Main Thread or the Swing Event Dispatch Thread is not recommended. *

*
Backpressure:
*
The operator consumes the source {@code Flowable} in an bounded manner (up to bufferSize * outstanding request amount for items).
*
Scheduler:
*
{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.15 - experimental * @param onNext the callback action for each source value * @param onError the callback action for an error event * @param onComplete the callback action for the completion event. * @param bufferSize the size of the buffer * @since 2.2 */ @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final void blockingSubscribe(Consumer onNext, Consumer onError, Action onComplete, int bufferSize) { FlowableBlockingSubscribe.subscribe(this, onNext, onError, onComplete, bufferSize); } /** * Subscribes to the source and calls the {@link Subscriber} methods on the current thread. *

* Note that calling this method will block the caller thread until the upstream terminates * normally, with an error or the {@code Subscriber} cancels the {@link Subscription} it receives via * {@link Subscriber#onSubscribe(Subscription)}. * Therefore, calling this method from special threads such as the * Android Main Thread or the Swing Event Dispatch Thread is not recommended. *

*
Backpressure:
*
The supplied {@code Subscriber} determines how backpressure is applied.
*
Scheduler:
*
{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.
*
* The cancellation and backpressure is composed through. * @param subscriber the subscriber to forward events and calls to in the current thread * @since 2.0 */ @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final void blockingSubscribe(Subscriber subscriber) { FlowableBlockingSubscribe.subscribe(this, subscriber); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping buffers, each containing {@code count} items. When the source * Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the * source Publisher. Note that if the source Publisher issues an onError notification the event is passed on * immediately without first emitting the buffer it is in the process of assembling. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and expects the source {@code Publisher} to honor it as * well, although not enforced; violation may lead to {@code MissingBackpressureException} somewhere * downstream.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param count * the maximum number of items in each buffer before it should be emitted * @return a Flowable that emits connected, non-overlapping buffers, each containing at most * {@code count} items from the source Publisher * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> buffer(int count) { return buffer(count, count); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits buffers every {@code skip} items, each containing {@code count} items. When the source * Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the * source Publisher. Note that if the source Publisher issues an onError notification the event is passed on * immediately without first emitting the buffer it is in the process of assembling. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and expects the source {@code Publisher} to honor it as * well, although not enforced; violation may lead to {@code MissingBackpressureException} somewhere * downstream.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param count * the maximum size of each buffer before it should be emitted * @param skip * how many items emitted by the source Publisher should be skipped before starting a new * buffer. Note that when {@code skip} and {@code count} are equal, this is the same operation as * {@link #buffer(int)}. * @return a Flowable that emits buffers for every {@code skip} item from the source Publisher and * containing at most {@code count} items * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> buffer(int count, int skip) { return buffer(count, skip, ArrayListSupplier.asCallable()); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits buffers every {@code skip} items, each containing {@code count} items. When the source * Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the * source Publisher. Note that if the source Publisher issues an onError notification the event is passed on * immediately without first emitting the buffer it is in the process of assembling. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and expects the source {@code Publisher} to honor it as * well, although not enforced; violation may lead to {@code MissingBackpressureException} somewhere * downstream.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param the collection subclass type to buffer into * @param count * the maximum size of each buffer before it should be emitted * @param skip * how many items emitted by the source Publisher should be skipped before starting a new * buffer. Note that when {@code skip} and {@code count} are equal, this is the same operation as * {@link #buffer(int)}. * @param bufferSupplier * a factory function that returns an instance of the collection subclass to be used and returned * as the buffer * @return a Flowable that emits buffers for every {@code skip} item from the source Publisher and * containing at most {@code count} items * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final > Flowable buffer(int count, int skip, Callable bufferSupplier) { ObjectHelper.verifyPositive(count, "count"); ObjectHelper.verifyPositive(skip, "skip"); ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null"); return RxJavaPlugins.onAssembly(new FlowableBuffer(this, count, skip, bufferSupplier)); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping buffers, each containing {@code count} items. When the source * Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the * source Publisher. Note that if the source Publisher issues an onError notification the event is passed on * immediately without first emitting the buffer it is in the process of assembling. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and expects the source {@code Publisher} to honor it as * well, although not enforced; violation may lead to {@code MissingBackpressureException} somewhere * downstream.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param the collection subclass type to buffer into * @param count * the maximum number of items in each buffer before it should be emitted * @param bufferSupplier * a factory function that returns an instance of the collection subclass to be used and returned * as the buffer * @return a Flowable that emits connected, non-overlapping buffers, each containing at most * {@code count} items from the source Publisher * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final > Flowable buffer(int count, Callable bufferSupplier) { return buffer(count, count, bufferSupplier); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher starts a new buffer periodically, as determined by the {@code timeskip} argument. It emits * each buffer after a fixed timespan, specified by the {@code timespan} argument. When the source * Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the * source Publisher. Note that if the source Publisher issues an onError notification the event is passed on * immediately without first emitting the buffer it is in the process of assembling. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE} * upstream and does not obey downstream requests.
*
Scheduler:
*
This version of {@code buffer} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timespan * the period of time each buffer collects items before it is emitted * @param timeskip * the period of time after which a new buffer will be created * @param unit * the unit of time that applies to the {@code timespan} and {@code timeskip} arguments * @return a Flowable that emits new buffers of items emitted by the source Publisher periodically after * a fixed timespan has elapsed * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable> buffer(long timespan, long timeskip, TimeUnit unit) { return buffer(timespan, timeskip, unit, Schedulers.computation(), ArrayListSupplier.asCallable()); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher starts a new buffer periodically, as determined by the {@code timeskip} argument, and on the * specified {@code scheduler}. It emits each buffer after a fixed timespan, specified by the * {@code timespan} argument. When the source Publisher completes, the resulting Publisher emits the current buffer * and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError * notification the event is passed on immediately without first emitting the buffer it is in the process of * assembling. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE} * upstream and does not obey downstream requests.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timespan * the period of time each buffer collects items before it is emitted * @param timeskip * the period of time after which a new buffer will be created * @param unit * the unit of time that applies to the {@code timespan} and {@code timeskip} arguments * @param scheduler * the {@link Scheduler} to use when determining the end and start of a buffer * @return a Flowable that emits new buffers of items emitted by the source Publisher periodically after * a fixed timespan has elapsed * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable> buffer(long timespan, long timeskip, TimeUnit unit, Scheduler scheduler) { return buffer(timespan, timeskip, unit, scheduler, ArrayListSupplier.asCallable()); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher starts a new buffer periodically, as determined by the {@code timeskip} argument, and on the * specified {@code scheduler}. It emits each buffer after a fixed timespan, specified by the * {@code timespan} argument. When the source Publisher completes, the resulting Publisher emits the current buffer * and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError * notification the event is passed on immediately without first emitting the buffer it is in the process of * assembling. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE} * upstream and does not obey downstream requests.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param the collection subclass type to buffer into * @param timespan * the period of time each buffer collects items before it is emitted * @param timeskip * the period of time after which a new buffer will be created * @param unit * the unit of time that applies to the {@code timespan} and {@code timeskip} arguments * @param scheduler * the {@link Scheduler} to use when determining the end and start of a buffer * @param bufferSupplier * a factory function that returns an instance of the collection subclass to be used and returned * as the buffer * @return a Flowable that emits new buffers of items emitted by the source Publisher periodically after * a fixed timespan has elapsed * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final > Flowable buffer(long timespan, long timeskip, TimeUnit unit, Scheduler scheduler, Callable bufferSupplier) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null"); return RxJavaPlugins.onAssembly(new FlowableBufferTimed(this, timespan, timeskip, unit, scheduler, bufferSupplier, Integer.MAX_VALUE, false)); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the * {@code timespan} argument. When the source Publisher completes, the resulting Publisher emits the current buffer * and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError * notification the event is passed on immediately without first emitting the buffer it is in the process of * assembling. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE} * upstream and does not obey downstream requests.
*
Scheduler:
*
This version of {@code buffer} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timespan * the period of time each buffer collects items before it is emitted and replaced with a new * buffer * @param unit * the unit of time that applies to the {@code timespan} argument * @return a Flowable that emits connected, non-overlapping buffers of items emitted by the source * Publisher within a fixed duration * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable> buffer(long timespan, TimeUnit unit) { return buffer(timespan, unit, Schedulers.computation(), Integer.MAX_VALUE); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the * {@code timespan} argument or a maximum size specified by the {@code count} argument (whichever is reached * first). When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the * notification from the source Publisher. Note that if the source Publisher issues an onError notification the event * is passed on immediately without first emitting the buffer it is in the process of assembling. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE} * upstream and does not obey downstream requests.
*
Scheduler:
*
This version of {@code buffer} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timespan * the period of time each buffer collects items before it is emitted and replaced with a new * buffer * @param unit * the unit of time which applies to the {@code timespan} argument * @param count * the maximum size of each buffer before it is emitted * @return a Flowable that emits connected, non-overlapping buffers of items emitted by the source * Publisher, after a fixed duration or when the buffer reaches maximum capacity (whichever occurs * first) * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable> buffer(long timespan, TimeUnit unit, int count) { return buffer(timespan, unit, Schedulers.computation(), count); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the * {@code timespan} argument as measured on the specified {@code scheduler}, or a maximum size specified by * the {@code count} argument (whichever is reached first). When the source Publisher completes, the resulting * Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the * source Publisher issues an onError notification the event is passed on immediately without first emitting the * buffer it is in the process of assembling. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE} * upstream and does not obey downstream requests.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timespan * the period of time each buffer collects items before it is emitted and replaced with a new * buffer * @param unit * the unit of time which applies to the {@code timespan} argument * @param scheduler * the {@link Scheduler} to use when determining the end and start of a buffer * @param count * the maximum size of each buffer before it is emitted * @return a Flowable that emits connected, non-overlapping buffers of items emitted by the source * Publisher after a fixed duration or when the buffer reaches maximum capacity (whichever occurs * first) * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable> buffer(long timespan, TimeUnit unit, Scheduler scheduler, int count) { return buffer(timespan, unit, scheduler, count, ArrayListSupplier.asCallable(), false); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the * {@code timespan} argument as measured on the specified {@code scheduler}, or a maximum size specified by * the {@code count} argument (whichever is reached first). When the source Publisher completes, the resulting * Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the * source Publisher issues an onError notification the event is passed on immediately without first emitting the * buffer it is in the process of assembling. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE} * upstream and does not obey downstream requests.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param the collection subclass type to buffer into * @param timespan * the period of time each buffer collects items before it is emitted and replaced with a new * buffer * @param unit * the unit of time which applies to the {@code timespan} argument * @param scheduler * the {@link Scheduler} to use when determining the end and start of a buffer * @param count * the maximum size of each buffer before it is emitted * @param bufferSupplier * a factory function that returns an instance of the collection subclass to be used and returned * as the buffer * @param restartTimerOnMaxSize if true the time window is restarted when the max capacity of the current buffer * is reached * @return a Flowable that emits connected, non-overlapping buffers of items emitted by the source * Publisher after a fixed duration or when the buffer reaches maximum capacity (whichever occurs * first) * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final > Flowable buffer( long timespan, TimeUnit unit, Scheduler scheduler, int count, Callable bufferSupplier, boolean restartTimerOnMaxSize) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null"); ObjectHelper.verifyPositive(count, "count"); return RxJavaPlugins.onAssembly(new FlowableBufferTimed(this, timespan, timespan, unit, scheduler, bufferSupplier, count, restartTimerOnMaxSize)); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the * {@code timespan} argument and on the specified {@code scheduler}. When the source Publisher completes, the * resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that * if the source Publisher issues an onError notification the event is passed on immediately without first emitting * the buffer it is in the process of assembling. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE} * upstream and does not obey downstream requests.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timespan * the period of time each buffer collects items before it is emitted and replaced with a new * buffer * @param unit * the unit of time which applies to the {@code timespan} argument * @param scheduler * the {@link Scheduler} to use when determining the end and start of a buffer * @return a Flowable that emits connected, non-overlapping buffers of items emitted by the source * Publisher within a fixed duration * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable> buffer(long timespan, TimeUnit unit, Scheduler scheduler) { return buffer(timespan, unit, scheduler, Integer.MAX_VALUE, ArrayListSupplier.asCallable(), false); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits buffers that it creates when the specified {@code openingIndicator} Publisher emits an * item, and closes when the Publisher returned from {@code closingIndicator} emits an item. If any of the source * Publisher, {@code openingIndicator} or {@code closingIndicator} issues an onError notification the event is passed * on immediately without first emitting the buffer it is in the process of assembling. *

* *

*
Backpressure:
*
This operator does not support backpressure as it is instead controlled by the given Publishers and * buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey downstream requests.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the buffer-opening Publisher * @param the element type of the individual buffer-closing Publishers * @param openingIndicator * the Publisher that, when it emits an item, causes a new buffer to be created * @param closingIndicator * the {@link Function} that is used to produce a Publisher for every buffer created. When this * Publisher emits an item, the associated buffer is emitted. * @return a Flowable that emits buffers, containing items from the source Publisher, that are created * and closed when the specified Publishers emit items * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> buffer( Flowable openingIndicator, Function> closingIndicator) { return buffer(openingIndicator, closingIndicator, ArrayListSupplier.asCallable()); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits buffers that it creates when the specified {@code openingIndicator} Publisher emits an * item, and closes when the Publisher returned from {@code closingIndicator} emits an item. If any of the source * Publisher, {@code openingIndicator} or {@code closingIndicator} issues an onError notification the event is passed * on immediately without first emitting the buffer it is in the process of assembling. *

* *

*
Backpressure:
*
This operator does not support backpressure as it is instead controlled by the given Publishers and * buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey downstream requests.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param the collection subclass type to buffer into * @param the element type of the buffer-opening Publisher * @param the element type of the individual buffer-closing Publishers * @param openingIndicator * the Publisher that, when it emits an item, causes a new buffer to be created * @param closingIndicator * the {@link Function} that is used to produce a Publisher for every buffer created. When this * Publisher emits an item, the associated buffer is emitted. * @param bufferSupplier * a factory function that returns an instance of the collection subclass to be used and returned * as the buffer * @return a Flowable that emits buffers, containing items from the source Publisher, that are created * and closed when the specified Publishers emit items * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final > Flowable buffer( Flowable openingIndicator, Function> closingIndicator, Callable bufferSupplier) { ObjectHelper.requireNonNull(openingIndicator, "openingIndicator is null"); ObjectHelper.requireNonNull(closingIndicator, "closingIndicator is null"); ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null"); return RxJavaPlugins.onAssembly(new FlowableBufferBoundary(this, openingIndicator, closingIndicator, bufferSupplier)); } /** * Returns a Flowable that emits non-overlapping buffered items from the source Publisher each time the * specified boundary Publisher emits an item. *

* *

* Completion of either the source or the boundary Publisher causes the returned Publisher to emit the * latest buffer and complete. If either the source Publisher or the boundary Publisher issues an onError notification * the event is passed on immediately without first emitting the buffer it is in the process of assembling. *

*
Backpressure:
*
This operator does not support backpressure as it is instead controlled by the {@code Publisher} * {@code boundary} and buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey * downstream requests.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the boundary value type (ignored) * @param boundaryIndicator * the boundary Publisher * @return a Flowable that emits buffered items from the source Publisher when the boundary Publisher * emits an item * @see #buffer(Publisher, int) * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> buffer(Publisher boundaryIndicator) { return buffer(boundaryIndicator, ArrayListSupplier.asCallable()); } /** * Returns a Flowable that emits non-overlapping buffered items from the source Publisher each time the * specified boundary Publisher emits an item. *

* *

* Completion of either the source or the boundary Publisher causes the returned Publisher to emit the * latest buffer and complete. If either the source Publisher or the boundary Publisher issues an onError notification * the event is passed on immediately without first emitting the buffer it is in the process of assembling. *

*
Backpressure:
*
This operator does not support backpressure as it is instead controlled by the {@code Publisher} * {@code boundary} and buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey * downstream requests.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the boundary value type (ignored) * @param boundaryIndicator * the boundary Publisher * @param initialCapacity * the initial capacity of each buffer chunk * @return a Flowable that emits buffered items from the source Publisher when the boundary Publisher * emits an item * @see ReactiveX operators documentation: Buffer * @see #buffer(Publisher) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> buffer(Publisher boundaryIndicator, final int initialCapacity) { ObjectHelper.verifyPositive(initialCapacity, "initialCapacity"); return buffer(boundaryIndicator, Functions.createArrayList(initialCapacity)); } /** * Returns a Flowable that emits non-overlapping buffered items from the source Publisher each time the * specified boundary Publisher emits an item. *

* *

* Completion of either the source or the boundary Publisher causes the returned Publisher to emit the * latest buffer and complete. If either the source Publisher or the boundary Publisher issues an onError notification * the event is passed on immediately without first emitting the buffer it is in the process of assembling. *

*
Backpressure:
*
This operator does not support backpressure as it is instead controlled by the {@code Publisher} * {@code boundary} and buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey * downstream requests.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param the collection subclass type to buffer into * @param * the boundary value type (ignored) * @param boundaryIndicator * the boundary Publisher * @param bufferSupplier * a factory function that returns an instance of the collection subclass to be used and returned * as the buffer * @return a Flowable that emits buffered items from the source Publisher when the boundary Publisher * emits an item * @see #buffer(Publisher, int) * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final > Flowable buffer(Publisher boundaryIndicator, Callable bufferSupplier) { ObjectHelper.requireNonNull(boundaryIndicator, "boundaryIndicator is null"); ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null"); return RxJavaPlugins.onAssembly(new FlowableBufferExactBoundary(this, boundaryIndicator, bufferSupplier)); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping buffers. It emits the current buffer and replaces it with a * new buffer whenever the Publisher produced by the specified {@code boundaryIndicatorSupplier} emits an item. *

* *

* If either the source {@code Publisher} or the boundary {@code Publisher} issues an {@code onError} notification the event is passed on * immediately without first emitting the buffer it is in the process of assembling. *

*
Backpressure:
*
This operator does not support backpressure as it is instead controlled by the given Publishers and * buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey downstream requests.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the boundary-providing Publisher * @param boundaryIndicatorSupplier * a {@link Callable} that produces a Publisher that governs the boundary between buffers. * Whenever the supplied {@code Publisher} emits an item, {@code buffer} emits the current buffer and * begins to fill a new one * @return a Flowable that emits a connected, non-overlapping buffer of items from the source Publisher * each time the Publisher created with the {@code closingIndicator} argument emits an item * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> buffer(Callable> boundaryIndicatorSupplier) { return buffer(boundaryIndicatorSupplier, ArrayListSupplier.asCallable()); } /** * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping buffers. It emits the current buffer and replaces it with a * new buffer whenever the Publisher produced by the specified {@code boundaryIndicatorSupplier} emits an item. *

* *

* If either the source {@code Publisher} or the boundary {@code Publisher} issues an {@code onError} notification the event is passed on * immediately without first emitting the buffer it is in the process of assembling. *

*
Backpressure:
*
This operator does not support backpressure as it is instead controlled by the given Publishers and * buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey downstream requests.
*
Scheduler:
*
This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param the collection subclass type to buffer into * @param the value type of the boundary-providing Publisher * @param boundaryIndicatorSupplier * a {@link Callable} that produces a Publisher that governs the boundary between buffers. * Whenever the supplied {@code Publisher} emits an item, {@code buffer} emits the current buffer and * begins to fill a new one * @param bufferSupplier * a factory function that returns an instance of the collection subclass to be used and returned * as the buffer * @return a Flowable that emits a connected, non-overlapping buffer of items from the source Publisher * each time the Publisher created with the {@code closingIndicator} argument emits an item * @see ReactiveX operators documentation: Buffer */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final > Flowable buffer(Callable> boundaryIndicatorSupplier, Callable bufferSupplier) { ObjectHelper.requireNonNull(boundaryIndicatorSupplier, "boundaryIndicatorSupplier is null"); ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null"); return RxJavaPlugins.onAssembly(new FlowableBufferBoundarySupplier(this, boundaryIndicatorSupplier, bufferSupplier)); } /** * Returns a Flowable that subscribes to this Publisher lazily, caches all of its events * and replays them, in the same order as received, to all the downstream subscribers. *

* *

* This is useful when you want a Publisher to cache responses and you can't control the * subscribe/cancel behavior of all the {@link Subscriber}s. *

* The operator subscribes only when the first downstream subscriber subscribes and maintains * a single subscription towards this Publisher. In contrast, the operator family of {@link #replay()} * that return a {@link ConnectableFlowable} require an explicit call to {@link ConnectableFlowable#connect()}. *

* Note: You sacrifice the ability to cancel the origin when you use the {@code cache} * Subscriber so be careful not to use this Subscriber on Publishers that emit an infinite or very large number * of items that will use up memory. * A possible workaround is to apply `takeUntil` with a predicate or * another source before (and perhaps after) the application of cache(). *


     * AtomicBoolean shouldStop = new AtomicBoolean();
     *
     * source.takeUntil(v -> shouldStop.get())
     *       .cache()
     *       .takeUntil(v -> shouldStop.get())
     *       .subscribe(...);
     * 
* Since the operator doesn't allow clearing the cached values either, the possible workaround is * to forget all references to it via {@link #onTerminateDetach()} applied along with the previous * workaround: *

     * AtomicBoolean shouldStop = new AtomicBoolean();
     *
     * source.takeUntil(v -> shouldStop.get())
     *       .onTerminateDetach()
     *       .cache()
     *       .takeUntil(v -> shouldStop.get())
     *       .onTerminateDetach()
     *       .subscribe(...);
     * 
*
*
Backpressure:
*
The operator consumes this Publisher in an unbounded fashion but respects the backpressure * of each downstream Subscriber individually.
*
Scheduler:
*
{@code cache} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Flowable that, when first subscribed to, caches all of its items and notifications for the * benefit of subsequent subscribers * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable cache() { return cacheWithInitialCapacity(16); } /** * Returns a Flowable that subscribes to this Publisher lazily, caches all of its events * and replays them, in the same order as received, to all the downstream subscribers. *

* *

* This is useful when you want a Publisher to cache responses and you can't control the * subscribe/cancel behavior of all the {@link Subscriber}s. *

* The operator subscribes only when the first downstream subscriber subscribes and maintains * a single subscription towards this Publisher. In contrast, the operator family of {@link #replay()} * that return a {@link ConnectableFlowable} require an explicit call to {@link ConnectableFlowable#connect()}. *

* Note: You sacrifice the ability to cancel the origin when you use the {@code cache} * Subscriber so be careful not to use this Subscriber on Publishers that emit an infinite or very large number * of items that will use up memory. * A possible workaround is to apply `takeUntil` with a predicate or * another source before (and perhaps after) the application of cache(). *


     * AtomicBoolean shouldStop = new AtomicBoolean();
     *
     * source.takeUntil(v -> shouldStop.get())
     *       .cache()
     *       .takeUntil(v -> shouldStop.get())
     *       .subscribe(...);
     * 
* Since the operator doesn't allow clearing the cached values either, the possible workaround is * to forget all references to it via {@link #onTerminateDetach()} applied along with the previous * workaround: *

     * AtomicBoolean shouldStop = new AtomicBoolean();
     *
     * source.takeUntil(v -> shouldStop.get())
     *       .onTerminateDetach()
     *       .cache()
     *       .takeUntil(v -> shouldStop.get())
     *       .onTerminateDetach()
     *       .subscribe(...);
     * 
*
*
Backpressure:
*
The operator consumes this Publisher in an unbounded fashion but respects the backpressure * of each downstream Subscriber individually.
*
Scheduler:
*
{@code cacheWithInitialCapacity} does not operate by default on a particular {@link Scheduler}.
*
*

* Note: The capacity hint is not an upper bound on cache size. For that, consider * {@link #replay(int)} in combination with {@link ConnectableFlowable#autoConnect()} or similar. * * @param initialCapacity hint for number of items to cache (for optimizing underlying data structure) * @return a Flowable that, when first subscribed to, caches all of its items and notifications for the * benefit of subsequent subscribers * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable cacheWithInitialCapacity(int initialCapacity) { ObjectHelper.verifyPositive(initialCapacity, "initialCapacity"); return RxJavaPlugins.onAssembly(new FlowableCache(this, initialCapacity)); } /** * Returns a Flowable that emits the items emitted by the source Publisher, converted to the specified * type. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code cast} does not operate by default on a particular {@link Scheduler}.
*
* * @param the output value type cast to * @param clazz * the target class type that {@code cast} will cast the items emitted by the source Publisher * into before emitting them from the resulting Publisher * @return a Flowable that emits each item from the source Publisher after converting it to the * specified type * @see ReactiveX operators documentation: Map */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable cast(final Class clazz) { ObjectHelper.requireNonNull(clazz, "clazz is null"); return map(Functions.castFunction(clazz)); } /** * Collects items emitted by the finite source Publisher into a single mutable data structure and returns * a Single that emits this structure. *

* *

* This is a simplified version of {@code reduce} that does not need to return the state on each pass. *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulator object to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
This operator does not support backpressure because by intent it will receive all values and reduce * them to a single {@code onNext}.
*
Scheduler:
*
{@code collect} does not operate by default on a particular {@link Scheduler}.
*
* * @param the accumulator and output type * @param initialItemSupplier * the mutable data structure that will collect the items * @param collector * a function that accepts the {@code state} and an emitted item, and modifies {@code state} * accordingly * @return a Single that emits the result of collecting the values emitted by the source Publisher * into a single mutable data structure * @see ReactiveX operators documentation: Reduce */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single collect(Callable initialItemSupplier, BiConsumer collector) { ObjectHelper.requireNonNull(initialItemSupplier, "initialItemSupplier is null"); ObjectHelper.requireNonNull(collector, "collector is null"); return RxJavaPlugins.onAssembly(new FlowableCollectSingle(this, initialItemSupplier, collector)); } /** * Collects items emitted by the finite source Publisher into a single mutable data structure and returns * a Single that emits this structure. *

* *

* This is a simplified version of {@code reduce} that does not need to return the state on each pass. *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulator object to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
This operator does not support backpressure because by intent it will receive all values and reduce * them to a single {@code onNext}.
*
Scheduler:
*
{@code collectInto} does not operate by default on a particular {@link Scheduler}.
*
* * @param the accumulator and output type * @param initialItem * the mutable data structure that will collect the items * @param collector * a function that accepts the {@code state} and an emitted item, and modifies {@code state} * accordingly * @return a Single that emits the result of collecting the values emitted by the source Publisher * into a single mutable data structure * @see ReactiveX operators documentation: Reduce */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single collectInto(final U initialItem, BiConsumer collector) { ObjectHelper.requireNonNull(initialItem, "initialItem is null"); return collect(Functions.justCallable(initialItem), collector); } /** * Transform a Publisher by applying a particular Transformer function to it. *

* This method operates on the Publisher itself whereas {@link #lift} operates on the Publisher's * Subscribers or Subscribers. *

* If the operator you are creating is designed to act on the individual items emitted by a source * Publisher, use {@link #lift}. If your operator is designed to transform the source Publisher as a whole * (for instance, by applying a particular set of existing RxJava operators to it) use {@code compose}. *

*
Backpressure:
*
The operator itself doesn't interfere with the backpressure behavior which only depends * on what kind of {@code Publisher} the transformer returns.
*
Scheduler:
*
{@code compose} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the output Publisher * @param composer implements the function that transforms the source Publisher * @return the source Publisher, transformed by the transformer function * @see RxJava wiki: Implementing Your Own Operators */ @SuppressWarnings("unchecked") @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable compose(FlowableTransformer composer) { return fromPublisher(((FlowableTransformer) ObjectHelper.requireNonNull(composer, "composer is null")).apply(this)); } /** * Returns a new Flowable that emits items resulting from applying a function that you supply to each item * emitted by the source Publisher, where that function returns a Publisher, and then emitting the items * that result from concatenating those resulting Publishers. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. Both this and the inner {@code Publisher}s are * expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will * signal a {@code MissingBackpressureException}. If any of the inner {@code Publisher}s doesn't honor * backpressure, that may throw an {@code IllegalStateException} when that * {@code Publisher} completes.
*
Scheduler:
*
{@code concatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the type of the inner Publisher sources and thus the output type * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @return a Flowable that emits the result of applying the transformation function to each item emitted * by the source Publisher and concatenating the Publishers obtained from this transformation * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMap(Function> mapper) { return concatMap(mapper, 2); } /** * Returns a new Flowable that emits items resulting from applying a function that you supply to each item * emitted by the source Publisher, where that function returns a Publisher, and then emitting the items * that result from concatenating those resulting Publishers. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. Both this and the inner {@code Publisher}s are * expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will * signal a {@code MissingBackpressureException}. If any of the inner {@code Publisher}s doesn't honor * backpressure, that may throw an {@code IllegalStateException} when that * {@code Publisher} completes.
*
Scheduler:
*
{@code concatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the type of the inner Publisher sources and thus the output type * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @param prefetch * the number of elements to prefetch from the current Flowable * @return a Flowable that emits the result of applying the transformation function to each item emitted * by the source Publisher and concatenating the Publishers obtained from this transformation * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMap(Function> mapper, int prefetch) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(prefetch, "prefetch"); if (this instanceof ScalarCallable) { @SuppressWarnings("unchecked") T v = ((ScalarCallable)this).call(); if (v == null) { return empty(); } return FlowableScalarXMap.scalarXMap(v, mapper); } return RxJavaPlugins.onAssembly(new FlowableConcatMap(this, mapper, prefetch, ErrorMode.IMMEDIATE)); } /** * Maps the upstream items into {@link CompletableSource}s and subscribes to them one after the * other completes. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapCompletable} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param mapper the function called with the upstream item and should return * a {@code CompletableSource} to become the next source to * be subscribed to * @return a new Completable instance * @see #concatMapCompletableDelayError(Function) * @since 2.2 */ @CheckReturnValue @SchedulerSupport(SchedulerSupport.NONE) @BackpressureSupport(BackpressureKind.FULL) public final Completable concatMapCompletable(Function mapper) { return concatMapCompletable(mapper, 2); } /** * Maps the upstream items into {@link CompletableSource}s and subscribes to them one after the * other completes. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapCompletable} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param mapper the function called with the upstream item and should return * a {@code CompletableSource} to become the next source to * be subscribed to * @param prefetch The number of upstream items to prefetch so that fresh items are * ready to be mapped when a previous {@code CompletableSource} terminates. * The operator replenishes after half of the prefetch amount has been consumed * and turned into {@code CompletableSource}s. * @return a new Completable instance * @see #concatMapCompletableDelayError(Function, boolean, int) * @since 2.2 */ @CheckReturnValue @NonNull @SchedulerSupport(SchedulerSupport.NONE) @BackpressureSupport(BackpressureKind.FULL) public final Completable concatMapCompletable(Function mapper, int prefetch) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapCompletable(this, mapper, ErrorMode.IMMEDIATE, prefetch)); } /** * Maps the upstream items into {@link CompletableSource}s and subscribes to them one after the * other terminates, delaying all errors till both this {@code Flowable} and all * inner {@code CompletableSource}s terminate. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapCompletableDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param mapper the function called with the upstream item and should return * a {@code CompletableSource} to become the next source to * be subscribed to * @return a new Completable instance * @see #concatMapCompletable(Function, int) * @since 2.2 */ @CheckReturnValue @SchedulerSupport(SchedulerSupport.NONE) @BackpressureSupport(BackpressureKind.FULL) public final Completable concatMapCompletableDelayError(Function mapper) { return concatMapCompletableDelayError(mapper, true, 2); } /** * Maps the upstream items into {@link CompletableSource}s and subscribes to them one after the * other terminates, optionally delaying all errors till both this {@code Flowable} and all * inner {@code CompletableSource}s terminate. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapCompletableDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param mapper the function called with the upstream item and should return * a {@code CompletableSource} to become the next source to * be subscribed to * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the * inner {@code CompletableSource}s are delayed until all * of them terminate. If {@code false}, an error from this * {@code Flowable} is delayed until the current inner * {@code CompletableSource} terminates and only then is * it emitted to the downstream. * @return a new Completable instance * @see #concatMapCompletable(Function) * @since 2.2 */ @CheckReturnValue @SchedulerSupport(SchedulerSupport.NONE) @BackpressureSupport(BackpressureKind.FULL) public final Completable concatMapCompletableDelayError(Function mapper, boolean tillTheEnd) { return concatMapCompletableDelayError(mapper, tillTheEnd, 2); } /** * Maps the upstream items into {@link CompletableSource}s and subscribes to them one after the * other terminates, optionally delaying all errors till both this {@code Flowable} and all * inner {@code CompletableSource}s terminate. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapCompletableDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param mapper the function called with the upstream item and should return * a {@code CompletableSource} to become the next source to * be subscribed to * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the * inner {@code CompletableSource}s are delayed until all * of them terminate. If {@code false}, an error from this * {@code Flowable} is delayed until the current inner * {@code CompletableSource} terminates and only then is * it emitted to the downstream. * @param prefetch The number of upstream items to prefetch so that fresh items are * ready to be mapped when a previous {@code CompletableSource} terminates. * The operator replenishes after half of the prefetch amount has been consumed * and turned into {@code CompletableSource}s. * @return a new Completable instance * @see #concatMapCompletable(Function, int) * @since 2.2 */ @CheckReturnValue @NonNull @SchedulerSupport(SchedulerSupport.NONE) @BackpressureSupport(BackpressureKind.FULL) public final Completable concatMapCompletableDelayError(Function mapper, boolean tillTheEnd, int prefetch) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapCompletable(this, mapper, tillTheEnd ? ErrorMode.END : ErrorMode.BOUNDARY, prefetch)); } /** * Maps each of the items into a Publisher, subscribes to them one after the other, * one at a time and emits their values in order * while delaying any error from either this or any of the inner Publishers * till all of them terminate. * *

*
Backpressure:
*
The operator honors backpressure from downstream. Both this and the inner {@code Publisher}s are * expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will * signal a {@code MissingBackpressureException}. If any of the inner {@code Publisher}s doesn't honor * backpressure, that may throw an {@code IllegalStateException} when that * {@code Publisher} completes.
*
Scheduler:
*
{@code concatMapDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the result value type * @param mapper the function that maps the items of this Publisher into the inner Publishers. * @return the new Publisher instance with the concatenation behavior */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapDelayError(Function> mapper) { return concatMapDelayError(mapper, 2, true); } /** * Maps each of the items into a Publisher, subscribes to them one after the other, * one at a time and emits their values in order * while delaying any error from either this or any of the inner Publishers * till all of them terminate. * *
*
Backpressure:
*
The operator honors backpressure from downstream. Both this and the inner {@code Publisher}s are * expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will * signal a {@code MissingBackpressureException}. If any of the inner {@code Publisher}s doesn't honor * backpressure, that may throw an {@code IllegalStateException} when that * {@code Publisher} completes.
*
Scheduler:
*
{@code concatMapDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the result value type * @param mapper the function that maps the items of this Publisher into the inner Publishers. * @param prefetch * the number of elements to prefetch from the current Flowable * @param tillTheEnd * if true, all errors from the outer and inner Publisher sources are delayed until the end, * if false, an error from the main source is signaled when the current Publisher source terminates * @return the new Publisher instance with the concatenation behavior */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapDelayError(Function> mapper, int prefetch, boolean tillTheEnd) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(prefetch, "prefetch"); if (this instanceof ScalarCallable) { @SuppressWarnings("unchecked") T v = ((ScalarCallable)this).call(); if (v == null) { return empty(); } return FlowableScalarXMap.scalarXMap(v, mapper); } return RxJavaPlugins.onAssembly(new FlowableConcatMap(this, mapper, prefetch, tillTheEnd ? ErrorMode.END : ErrorMode.BOUNDARY)); } /** * Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single * Publisher. *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * source Publishers. The operator buffers the values emitted by these Publishers and then drains them in * order, each one after the previous one completes. *

*
Backpressure:
*
Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources * are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param mapper the function that maps a sequence of values into a sequence of Publishers that will be * eagerly concatenated * @return the new Publisher instance with the specified concatenation behavior * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapEager(Function> mapper) { return concatMapEager(mapper, bufferSize(), bufferSize()); } /** * Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single * Publisher. *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * source Publishers. The operator buffers the values emitted by these Publishers and then drains them in * order, each one after the previous one completes. *

*
Backpressure:
*
Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources * are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param mapper the function that maps a sequence of values into a sequence of Publishers that will be * eagerly concatenated * @param maxConcurrency the maximum number of concurrent subscribed Publishers * @param prefetch hints about the number of expected values from each inner Publisher, must be positive * @return the new Publisher instance with the specified concatenation behavior * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapEager(Function> mapper, int maxConcurrency, int prefetch) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapEager(this, mapper, maxConcurrency, prefetch, ErrorMode.IMMEDIATE)); } /** * Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single * Publisher. *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * source Publishers. The operator buffers the values emitted by these Publishers and then drains them in * order, each one after the previous one completes. *

*
Backpressure:
*
Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources * are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param mapper the function that maps a sequence of values into a sequence of Publishers that will be * eagerly concatenated * @param tillTheEnd * if true, all errors from the outer and inner Publisher sources are delayed until the end, * if false, an error from the main source is signaled when the current Publisher source terminates * @return the new Publisher instance with the specified concatenation behavior * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapEagerDelayError(Function> mapper, boolean tillTheEnd) { return concatMapEagerDelayError(mapper, bufferSize(), bufferSize(), tillTheEnd); } /** * Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single * Publisher. *

* Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the * source Publishers. The operator buffers the values emitted by these Publishers and then drains them in * order, each one after the previous one completes. *

*
Backpressure:
*
Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources * are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* @param the value type * @param mapper the function that maps a sequence of values into a sequence of Publishers that will be * eagerly concatenated * @param maxConcurrency the maximum number of concurrent subscribed Publishers * @param prefetch * the number of elements to prefetch from each source Publisher * @param tillTheEnd * if true, exceptions from the current Flowable and all the inner Publishers are delayed until * all of them terminate, if false, exception from the current Flowable is delayed until the * currently running Publisher terminates * @return the new Publisher instance with the specified concatenation behavior * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapEagerDelayError(Function> mapper, int maxConcurrency, int prefetch, boolean tillTheEnd) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapEager(this, mapper, maxConcurrency, prefetch, tillTheEnd ? ErrorMode.END : ErrorMode.BOUNDARY)); } /** * Returns a Flowable that concatenate each item emitted by the source Publisher with the values in an * Iterable corresponding to that item that is generated by a selector. * *
*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s is * expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapIterable} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of item emitted by the resulting Publisher * @param mapper * a function that returns an Iterable sequence of values for when given an item emitted by the * source Publisher * @return a Flowable that emits the results of concatenating the items emitted by the source Publisher with * the values in the Iterables corresponding to those items, as generated by {@code collectionSelector} * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapIterable(Function> mapper) { return concatMapIterable(mapper, 2); } /** * Returns a Flowable that concatenate each item emitted by the source Publisher with the values in an * Iterable corresponding to that item that is generated by a selector. * *
*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s is * expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapIterable} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of item emitted by the resulting Publisher * @param mapper * a function that returns an Iterable sequence of values for when given an item emitted by the * source Publisher * @param prefetch * the number of elements to prefetch from the current Flowable * @return a Flowable that emits the results of concatenating the items emitted by the source Publisher with * the values in the Iterables corresponding to those items, as generated by {@code collectionSelector} * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapIterable(final Function> mapper, int prefetch) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableFlattenIterable(this, mapper, prefetch)); } /** * Maps the upstream items into {@link MaybeSource}s and subscribes to them one after the * other succeeds or completes, emits their success value if available or terminates immediately if * either this {@code Flowable} or the current inner {@code MaybeSource} fail. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure and honors * the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapMaybe} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the result type of the inner {@code MaybeSource}s * @param mapper the function called with the upstream item and should return * a {@code MaybeSource} to become the next source to * be subscribed to * @return a new Flowable instance * @see #concatMapMaybeDelayError(Function) * @see #concatMapMaybe(Function, int) * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapMaybe(Function> mapper) { return concatMapMaybe(mapper, 2); } /** * Maps the upstream items into {@link MaybeSource}s and subscribes to them one after the * other succeeds or completes, emits their success value if available or terminates immediately if * either this {@code Flowable} or the current inner {@code MaybeSource} fail. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure and honors * the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapMaybe} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the result type of the inner {@code MaybeSource}s * @param mapper the function called with the upstream item and should return * a {@code MaybeSource} to become the next source to * be subscribed to * @param prefetch The number of upstream items to prefetch so that fresh items are * ready to be mapped when a previous {@code MaybeSource} terminates. * The operator replenishes after half of the prefetch amount has been consumed * and turned into {@code MaybeSource}s. * @return a new Flowable instance * @see #concatMapMaybe(Function) * @see #concatMapMaybeDelayError(Function, boolean, int) * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapMaybe(Function> mapper, int prefetch) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapMaybe(this, mapper, ErrorMode.IMMEDIATE, prefetch)); } /** * Maps the upstream items into {@link MaybeSource}s and subscribes to them one after the * other terminates, emits their success value if available and delaying all errors * till both this {@code Flowable} and all inner {@code MaybeSource}s terminate. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure and honors * the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapMaybeDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the result type of the inner {@code MaybeSource}s * @param mapper the function called with the upstream item and should return * a {@code MaybeSource} to become the next source to * be subscribed to * @return a new Flowable instance * @see #concatMapMaybe(Function) * @see #concatMapMaybeDelayError(Function, boolean) * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapMaybeDelayError(Function> mapper) { return concatMapMaybeDelayError(mapper, true, 2); } /** * Maps the upstream items into {@link MaybeSource}s and subscribes to them one after the * other terminates, emits their success value if available and optionally delaying all errors * till both this {@code Flowable} and all inner {@code MaybeSource}s terminate. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure and honors * the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapMaybeDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the result type of the inner {@code MaybeSource}s * @param mapper the function called with the upstream item and should return * a {@code MaybeSource} to become the next source to * be subscribed to * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the * inner {@code MaybeSource}s are delayed until all * of them terminate. If {@code false}, an error from this * {@code Flowable} is delayed until the current inner * {@code MaybeSource} terminates and only then is * it emitted to the downstream. * @return a new Flowable instance * @see #concatMapMaybe(Function, int) * @see #concatMapMaybeDelayError(Function, boolean, int) * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapMaybeDelayError(Function> mapper, boolean tillTheEnd) { return concatMapMaybeDelayError(mapper, tillTheEnd, 2); } /** * Maps the upstream items into {@link MaybeSource}s and subscribes to them one after the * other terminates, emits their success value if available and optionally delaying all errors * till both this {@code Flowable} and all inner {@code MaybeSource}s terminate. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure and honors * the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapMaybeDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the result type of the inner {@code MaybeSource}s * @param mapper the function called with the upstream item and should return * a {@code MaybeSource} to become the next source to * be subscribed to * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the * inner {@code MaybeSource}s are delayed until all * of them terminate. If {@code false}, an error from this * {@code Flowable} is delayed until the current inner * {@code MaybeSource} terminates and only then is * it emitted to the downstream. * @param prefetch The number of upstream items to prefetch so that fresh items are * ready to be mapped when a previous {@code MaybeSource} terminates. * The operator replenishes after half of the prefetch amount has been consumed * and turned into {@code MaybeSource}s. * @return a new Flowable instance * @see #concatMapMaybe(Function, int) * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapMaybeDelayError(Function> mapper, boolean tillTheEnd, int prefetch) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapMaybe(this, mapper, tillTheEnd ? ErrorMode.END : ErrorMode.BOUNDARY, prefetch)); } /** * Maps the upstream items into {@link SingleSource}s and subscribes to them one after the * other succeeds, emits their success values or terminates immediately if * either this {@code Flowable} or the current inner {@code SingleSource} fail. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure and honors * the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapSingle} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the result type of the inner {@code SingleSource}s * @param mapper the function called with the upstream item and should return * a {@code SingleSource} to become the next source to * be subscribed to * @return a new Flowable instance * @see #concatMapSingleDelayError(Function) * @see #concatMapSingle(Function, int) * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapSingle(Function> mapper) { return concatMapSingle(mapper, 2); } /** * Maps the upstream items into {@link SingleSource}s and subscribes to them one after the * other succeeds, emits their success values or terminates immediately if * either this {@code Flowable} or the current inner {@code SingleSource} fail. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure and honors * the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapSingle} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the result type of the inner {@code SingleSource}s * @param mapper the function called with the upstream item and should return * a {@code SingleSource} to become the next source to * be subscribed to * @param prefetch The number of upstream items to prefetch so that fresh items are * ready to be mapped when a previous {@code SingleSource} terminates. * The operator replenishes after half of the prefetch amount has been consumed * and turned into {@code SingleSource}s. * @return a new Flowable instance * @see #concatMapSingle(Function) * @see #concatMapSingleDelayError(Function, boolean, int) * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapSingle(Function> mapper, int prefetch) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapSingle(this, mapper, ErrorMode.IMMEDIATE, prefetch)); } /** * Maps the upstream items into {@link SingleSource}s and subscribes to them one after the * other succeeds or fails, emits their success values and delays all errors * till both this {@code Flowable} and all inner {@code SingleSource}s terminate. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure and honors * the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapSingleDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the result type of the inner {@code SingleSource}s * @param mapper the function called with the upstream item and should return * a {@code SingleSource} to become the next source to * be subscribed to * @return a new Flowable instance * @see #concatMapSingle(Function) * @see #concatMapSingleDelayError(Function, boolean) * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapSingleDelayError(Function> mapper) { return concatMapSingleDelayError(mapper, true, 2); } /** * Maps the upstream items into {@link SingleSource}s and subscribes to them one after the * other succeeds or fails, emits their success values and optionally delays all errors * till both this {@code Flowable} and all inner {@code SingleSource}s terminate. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure and honors * the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapSingleDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the result type of the inner {@code SingleSource}s * @param mapper the function called with the upstream item and should return * a {@code SingleSource} to become the next source to * be subscribed to * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the * inner {@code SingleSource}s are delayed until all * of them terminate. If {@code false}, an error from this * {@code Flowable} is delayed until the current inner * {@code SingleSource} terminates and only then is * it emitted to the downstream. * @return a new Flowable instance * @see #concatMapSingle(Function, int) * @see #concatMapSingleDelayError(Function, boolean, int) * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapSingleDelayError(Function> mapper, boolean tillTheEnd) { return concatMapSingleDelayError(mapper, tillTheEnd, 2); } /** * Maps the upstream items into {@link SingleSource}s and subscribes to them one after the * other succeeds or fails, emits their success values and optionally delays errors * till both this {@code Flowable} and all inner {@code SingleSource}s terminate. *

* *

*
Backpressure:
*
The operator expects the upstream to support backpressure and honors * the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code concatMapSingleDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the result type of the inner {@code SingleSource}s * @param mapper the function called with the upstream item and should return * a {@code SingleSource} to become the next source to * be subscribed to * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the * inner {@code SingleSource}s are delayed until all * of them terminate. If {@code false}, an error from this * {@code Flowable} is delayed until the current inner * {@code SingleSource} terminates and only then is * it emitted to the downstream. * @param prefetch The number of upstream items to prefetch so that fresh items are * ready to be mapped when a previous {@code SingleSource} terminates. * The operator replenishes after half of the prefetch amount has been consumed * and turned into {@code SingleSource}s. * @return a new Flowable instance * @see #concatMapSingle(Function, int) * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatMapSingleDelayError(Function> mapper, boolean tillTheEnd, int prefetch) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowableConcatMapSingle(this, mapper, tillTheEnd ? ErrorMode.END : ErrorMode.BOUNDARY, prefetch)); } /** * Returns a Flowable that emits the items emitted from the current Publisher, then the next, one after * the other, without interleaving them. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. Both this and the {@code other} {@code Publisher}s * are expected to honor backpressure as well. If any of then violates this rule, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code concatWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param other * a Publisher to be concatenated after the current * @return a Flowable that emits items emitted by the two source Publishers, one after the other, * without interleaving them * @see ReactiveX operators documentation: Concat */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatWith(Publisher other) { ObjectHelper.requireNonNull(other, "other is null"); return concat(this, other); } /** * Returns a {@code Flowable} that emits the items from this {@code Flowable} followed by the success item or error event * of the other {@link SingleSource}. *

* *

*
Backpressure:
*
The operator supports backpressure and makes sure the success item of the other {@code SingleSource} * is only emitted when there is a demand for it.
*
Scheduler:
*
{@code concatWith} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.10 - experimental * @param other the SingleSource whose signal should be emitted after this {@code Flowable} completes normally. * @return the new Flowable instance * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatWith(@NonNull SingleSource other) { ObjectHelper.requireNonNull(other, "other is null"); return RxJavaPlugins.onAssembly(new FlowableConcatWithSingle(this, other)); } /** * Returns a {@code Flowable} that emits the items from this {@code Flowable} followed by the success item or terminal events * of the other {@link MaybeSource}. *

* *

*
Backpressure:
*
The operator supports backpressure and makes sure the success item of the other {@code MaybeSource} * is only emitted when there is a demand for it.
*
Scheduler:
*
{@code concatWith} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.10 - experimental * @param other the MaybeSource whose signal should be emitted after this Flowable completes normally. * @return the new Flowable instance * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatWith(@NonNull MaybeSource other) { ObjectHelper.requireNonNull(other, "other is null"); return RxJavaPlugins.onAssembly(new FlowableConcatWithMaybe(this, other)); } /** * Returns a {@code Flowable} that emits items from this {@code Flowable} and when it completes normally, the * other {@link CompletableSource} is subscribed to and the returned {@code Flowable} emits its terminal events. *

* *

*
Backpressure:
*
The operator does not interfere with backpressure between the current Flowable and the * downstream consumer (i.e., acts as pass-through). When the operator switches to the * {@code Completable}, backpressure is no longer present because {@code Completable} doesn't * have items to apply backpressure to.
*
Scheduler:
*
{@code concatWith} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.10 - experimental * @param other the {@code CompletableSource} to subscribe to once the current {@code Flowable} completes normally * @return the new Flowable instance * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable concatWith(@NonNull CompletableSource other) { ObjectHelper.requireNonNull(other, "other is null"); return RxJavaPlugins.onAssembly(new FlowableConcatWithCompletable(this, other)); } /** * Returns a Single that emits a Boolean that indicates whether the source Publisher emitted a * specified item. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code contains} does not operate by default on a particular {@link Scheduler}.
*
* * @param item * the item to search for in the emissions from the source Publisher * @return a Flowable that emits {@code true} if the specified item is emitted by the source Publisher, * or {@code false} if the source Publisher completes without emitting that item * @see ReactiveX operators documentation: Contains */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single contains(final Object item) { ObjectHelper.requireNonNull(item, "item is null"); return any(Functions.equalsWith(item)); } /** * Returns a Single that counts the total number of items emitted by the source Publisher and emits * this count as a 64-bit Long. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code count} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Single that emits a single item: the number of items emitted by the source Publisher as a * 64-bit Long item * @see ReactiveX operators documentation: Count */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single count() { return RxJavaPlugins.onAssembly(new FlowableCountSingle(this)); } /** * Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the * source Publisher that are followed by another item within a computed debounce duration. *

* *

* The delivery of the item happens on the thread of the first {@code onNext} or {@code onComplete} * signal of the generated {@code Publisher} sequence, * which if takes too long, a newer item may arrive from the upstream, causing the * generated sequence to get cancelled, which may also interrupt any downstream blocking operation * (yielding an {@code InterruptedException}). It is recommended processing items * that may take long time to be moved to another thread via {@link #observeOn} applied after * {@code debounce} itself. *

*
Backpressure:
*
This operator does not support backpressure as it uses the {@code debounceSelector} to mark * boundaries.
*
Scheduler:
*
This version of {@code debounce} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the debounce value type (ignored) * @param debounceIndicator * function to retrieve a sequence that indicates the throttle duration for each item * @return a Flowable that omits items emitted by the source Publisher that are followed by another item * within a computed debounce duration * @see ReactiveX operators documentation: Debounce * @see RxJava wiki: Backpressure */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable debounce(Function> debounceIndicator) { ObjectHelper.requireNonNull(debounceIndicator, "debounceIndicator is null"); return RxJavaPlugins.onAssembly(new FlowableDebounce(this, debounceIndicator)); } /** * Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the * source Publisher that are followed by newer items before a timeout value expires. The timer resets on * each emission. *

* Note: If items keep being emitted by the source Publisher faster than the timeout then no items * will be emitted by the resulting Publisher. *

* *

* Delivery of the item after the grace period happens on the {@code computation} {@code Scheduler}'s * {@code Worker} which if takes too long, a newer item may arrive from the upstream, causing the * {@code Worker}'s task to get disposed, which may also interrupt any downstream blocking operation * (yielding an {@code InterruptedException}). It is recommended processing items * that may take long time to be moved to another thread via {@link #observeOn} applied after * {@code debounce} itself. *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
{@code debounce} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timeout * the length of the window of time that must pass after the emission of an item from the source * Publisher in which that Publisher emits no items in order for the item to be emitted by the * resulting Publisher * @param unit * the unit of time for the specified {@code timeout} * @return a Flowable that filters out items from the source Publisher that are too quickly followed by * newer items * @see ReactiveX operators documentation: Debounce * @see RxJava wiki: Backpressure * @see #throttleWithTimeout(long, TimeUnit) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable debounce(long timeout, TimeUnit unit) { return debounce(timeout, unit, Schedulers.computation()); } /** * Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the * source Publisher that are followed by newer items before a timeout value expires on a specified * Scheduler. The timer resets on each emission. *

* Note: If items keep being emitted by the source Publisher faster than the timeout then no items * will be emitted by the resulting Publisher. *

* *

* Delivery of the item after the grace period happens on the given {@code Scheduler}'s * {@code Worker} which if takes too long, a newer item may arrive from the upstream, causing the * {@code Worker}'s task to get disposed, which may also interrupt any downstream blocking operation * (yielding an {@code InterruptedException}). It is recommended processing items * that may take long time to be moved to another thread via {@link #observeOn} applied after * {@code debounce} itself. *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timeout * the time each item has to be "the most recent" of those emitted by the source Publisher to * ensure that it's not dropped * @param unit * the unit of time for the specified {@code timeout} * @param scheduler * the {@link Scheduler} to use internally to manage the timers that handle the timeout for each * item * @return a Flowable that filters out items from the source Publisher that are too quickly followed by * newer items * @see ReactiveX operators documentation: Debounce * @see RxJava wiki: Backpressure * @see #throttleWithTimeout(long, TimeUnit, Scheduler) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable debounce(long timeout, TimeUnit unit, Scheduler scheduler) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableDebounceTimed(this, timeout, unit, scheduler)); } /** * Returns a Flowable that emits the items emitted by the source Publisher or a specified default item * if the source Publisher is empty. *

* *

*
Backpressure:
*
If the source {@code Publisher} is empty, this operator is guaranteed to honor backpressure from downstream. * If the source {@code Publisher} is non-empty, it is expected to honor backpressure as well; if the rule is violated, * a {@code MissingBackpressureException} may get signaled somewhere downstream. *
*
Scheduler:
*
{@code defaultIfEmpty} does not operate by default on a particular {@link Scheduler}.
*
* * @param defaultItem * the item to emit if the source Publisher emits no items * @return a Flowable that emits either the specified default item if the source Publisher emits no * items, or the items emitted by the source Publisher * @see ReactiveX operators documentation: DefaultIfEmpty */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable defaultIfEmpty(T defaultItem) { ObjectHelper.requireNonNull(defaultItem, "defaultItem is null"); return switchIfEmpty(just(defaultItem)); } /** * Returns a Flowable that delays the emissions of the source Publisher via another Publisher on a * per-item basis. *

* *

* Note: the resulting Publisher will immediately propagate any {@code onError} notification * from the source Publisher. *

*
Backpressure:
*
The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}. * All of the other {@code Publisher}s supplied by the function are consumed * in an unbounded manner (i.e., no backpressure applied to them).
*
Scheduler:
*
This version of {@code delay} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the item delay value type (ignored) * @param itemDelayIndicator * a function that returns a Publisher for each item emitted by the source Publisher, which is * then used to delay the emission of that item by the resulting Publisher until the Publisher * returned from {@code itemDelay} emits an item * @return a Flowable that delays the emissions of the source Publisher via another Publisher on a * per-item basis * @see ReactiveX operators documentation: Delay */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable delay(final Function> itemDelayIndicator) { ObjectHelper.requireNonNull(itemDelayIndicator, "itemDelayIndicator is null"); return flatMap(FlowableInternalHelper.itemDelay(itemDelayIndicator)); } /** * Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a * specified delay. Error notifications from the source Publisher are not delayed. *

* *

*
Backpressure:
*
The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.
*
Scheduler:
*
This version of {@code delay} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param delay * the delay to shift the source by * @param unit * the {@link TimeUnit} in which {@code period} is defined * @return the source Publisher shifted in time by the specified delay * @see ReactiveX operators documentation: Delay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable delay(long delay, TimeUnit unit) { return delay(delay, unit, Schedulers.computation(), false); } /** * Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a * specified delay. If {@code delayError} is true, error notifications will also be delayed. *

* *

*
Backpressure:
*
The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.
*
Scheduler:
*
This version of {@code delay} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param delay * the delay to shift the source by * @param unit * the {@link TimeUnit} in which {@code period} is defined * @param delayError * if true, the upstream exception is signaled with the given delay, after all preceding normal elements, * if false, the upstream exception is signaled immediately * @return the source Publisher shifted in time by the specified delay * @see ReactiveX operators documentation: Delay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable delay(long delay, TimeUnit unit, boolean delayError) { return delay(delay, unit, Schedulers.computation(), delayError); } /** * Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a * specified delay. Error notifications from the source Publisher are not delayed. *

* *

*
Backpressure:
*
The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param delay * the delay to shift the source by * @param unit * the time unit of {@code delay} * @param scheduler * the {@link Scheduler} to use for delaying * @return the source Publisher shifted in time by the specified delay * @see ReactiveX operators documentation: Delay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable delay(long delay, TimeUnit unit, Scheduler scheduler) { return delay(delay, unit, scheduler, false); } /** * Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a * specified delay. If {@code delayError} is true, error notifications will also be delayed. *

* *

*
Backpressure:
*
The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param delay * the delay to shift the source by * @param unit * the time unit of {@code delay} * @param scheduler * the {@link Scheduler} to use for delaying * @param delayError * if true, the upstream exception is signaled with the given delay, after all preceding normal elements, * if false, the upstream exception is signaled immediately * @return the source Publisher shifted in time by the specified delay * @see ReactiveX operators documentation: Delay */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable delay(long delay, TimeUnit unit, Scheduler scheduler, boolean delayError) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableDelay(this, Math.max(0L, delay), unit, scheduler, delayError)); } /** * Returns a Flowable that delays the subscription to and emissions from the source Publisher via another * Publisher on a per-item basis. *

* *

* Note: the resulting Publisher will immediately propagate any {@code onError} notification * from the source Publisher. *

*
Backpressure:
*
The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}. * All of the other {@code Publisher}s supplied by the functions are consumed * in an unbounded manner (i.e., no backpressure applied to them).
*
Scheduler:
*
This version of {@code delay} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the subscription delay value type (ignored) * @param * the item delay value type (ignored) * @param subscriptionIndicator * a function that returns a Publisher that triggers the subscription to the source Publisher * once it emits any item * @param itemDelayIndicator * a function that returns a Publisher for each item emitted by the source Publisher, which is * then used to delay the emission of that item by the resulting Publisher until the Publisher * returned from {@code itemDelay} emits an item * @return a Flowable that delays the subscription and emissions of the source Publisher via another * Publisher on a per-item basis * @see ReactiveX operators documentation: Delay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable delay(Publisher subscriptionIndicator, Function> itemDelayIndicator) { return delaySubscription(subscriptionIndicator).delay(itemDelayIndicator); } /** * Returns a Flowable that delays the subscription to this Publisher * until the other Publisher emits an element or completes normally. *
*
Backpressure:
*
The operator forwards the backpressure requests to this Publisher once * the subscription happens and requests Long.MAX_VALUE from the other Publisher
*
Scheduler:
*
This method does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the other Publisher, irrelevant * @param subscriptionIndicator the other Publisher that should trigger the subscription * to this Publisher. * @return a Flowable that delays the subscription to this Publisher * until the other Publisher emits an element or completes normally. * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable delaySubscription(Publisher subscriptionIndicator) { ObjectHelper.requireNonNull(subscriptionIndicator, "subscriptionIndicator is null"); return RxJavaPlugins.onAssembly(new FlowableDelaySubscriptionOther(this, subscriptionIndicator)); } /** * Returns a Flowable that delays the subscription to the source Publisher by a given amount of time. *

* *

*
Backpressure:
*
The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.
*
Scheduler:
*
This version of {@code delaySubscription} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param delay * the time to delay the subscription * @param unit * the time unit of {@code delay} * @return a Flowable that delays the subscription to the source Publisher by the given amount * @see ReactiveX operators documentation: Delay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable delaySubscription(long delay, TimeUnit unit) { return delaySubscription(delay, unit, Schedulers.computation()); } /** * Returns a Flowable that delays the subscription to the source Publisher by a given amount of time, * both waiting and subscribing on a given Scheduler. *

* *

*
Backpressure:
*
The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param delay * the time to delay the subscription * @param unit * the time unit of {@code delay} * @param scheduler * the Scheduler on which the waiting and subscription will happen * @return a Flowable that delays the subscription to the source Publisher by a given * amount, waiting and subscribing on the given Scheduler * @see ReactiveX operators documentation: Delay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable delaySubscription(long delay, TimeUnit unit, Scheduler scheduler) { return delaySubscription(timer(delay, unit, scheduler)); } /** * Returns a Flowable that reverses the effect of {@link #materialize materialize} by transforming the * {@link Notification} objects emitted by the source Publisher into the items or notifications they * represent. *

* *

* When the upstream signals an {@link Notification#createOnError(Throwable) onError} or * {@link Notification#createOnComplete() onComplete} item, the * returned Flowable cancels the flow and terminates with that type of terminal event: *


     * Flowable.just(createOnNext(1), createOnComplete(), createOnNext(2))
     * .doOnCancel(() -> System.out.println("Cancelled!"));
     * .dematerialize()
     * .test()
     * .assertResult(1);
     * 
* If the upstream signals {@code onError} or {@code onComplete} directly, the flow is terminated * with the same event. *

     * Flowable.just(createOnNext(1), createOnNext(2))
     * .dematerialize()
     * .test()
     * .assertResult(1, 2);
     * 
* If this behavior is not desired, the completion can be suppressed by applying {@link #concatWith(Publisher)} * with a {@link #never()} source. *
*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code dematerialize} does not operate by default on a particular {@link Scheduler}.
*
* * @param the output value type * @return a Flowable that emits the items and notifications embedded in the {@link Notification} objects * emitted by the source Publisher * @see ReactiveX operators documentation: Dematerialize * @see #dematerialize(Function) * @deprecated in 2.2.4; inherently type-unsafe as it overrides the output generic type. Use {@link #dematerialize(Function)} instead. */ @CheckReturnValue @SchedulerSupport(SchedulerSupport.NONE) @BackpressureSupport(BackpressureKind.PASS_THROUGH) @Deprecated @SuppressWarnings({ "unchecked", "rawtypes" }) public final Flowable dematerialize() { return RxJavaPlugins.onAssembly(new FlowableDematerialize(this, Functions.identity())); } /** * Returns a Flowable that reverses the effect of {@link #materialize materialize} by transforming the * {@link Notification} objects extracted from the source items via a selector function * into their respective {@code Subscriber} signal types. *

* *

* The intended use of the {@code selector} function is to perform a * type-safe identity mapping (see example) on a source that is already of type * {@code Notification}. The Java language doesn't allow * limiting instance methods to a certain generic argument shape, therefore, * a function is used to ensure the conversion remains type safe. *

* When the upstream signals an {@link Notification#createOnError(Throwable) onError} or * {@link Notification#createOnComplete() onComplete} item, the * returned Flowable cancels of the flow and terminates with that type of terminal event: *


     * Flowable.just(createOnNext(1), createOnComplete(), createOnNext(2))
     * .doOnCancel(() -> System.out.println("Canceled!"));
     * .dematerialize(notification -> notification)
     * .test()
     * .assertResult(1);
     * 
* If the upstream signals {@code onError} or {@code onComplete} directly, the flow is terminated * with the same event. *

     * Flowable.just(createOnNext(1), createOnNext(2))
     * .dematerialize(notification -> notification)
     * .test()
     * .assertResult(1, 2);
     * 
* If this behavior is not desired, the completion can be suppressed by applying {@link #concatWith(Publisher)} * with a {@link #never()} source. *
*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code dematerialize} does not operate by default on a particular {@link Scheduler}.
*
* * @param the output value type * @param selector function that returns the upstream item and should return a Notification to signal * the corresponding {@code Subscriber} event to the downstream. * @return a Flowable that emits the items and notifications embedded in the {@link Notification} objects * selected from the items emitted by the source Flowable * @see ReactiveX operators documentation: Dematerialize * @since 2.2.4 - experimental */ @Experimental @CheckReturnValue @NonNull @SchedulerSupport(SchedulerSupport.NONE) @BackpressureSupport(BackpressureKind.PASS_THROUGH) public final Flowable dematerialize(Function> selector) { ObjectHelper.requireNonNull(selector, "selector is null"); return RxJavaPlugins.onAssembly(new FlowableDematerialize(this, selector)); } /** * Returns a Flowable that emits all items emitted by the source Publisher that are distinct * based on {@link Object#equals(Object)} comparison. *

* *

* It is recommended the elements' class {@code T} in the flow overrides the default {@code Object.equals()} and {@link Object#hashCode()} to provide * a meaningful comparison between items as the default Java implementation only considers reference equivalence. *

* By default, {@code distinct()} uses an internal {@link java.util.HashSet} per Subscriber to remember * previously seen items and uses {@link java.util.Set#add(Object)} returning {@code false} as the * indicator for duplicates. *

* Note that this internal {@code HashSet} may grow unbounded as items won't be removed from it by * the operator. Therefore, using very long or infinite upstream (with very distinct elements) may lead * to {@code OutOfMemoryError}. *

* Customizing the retention policy can happen only by providing a custom {@link java.util.Collection} implementation * to the {@link #distinct(Function, Callable)} overload. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code distinct} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Flowable that emits only those items emitted by the source Publisher that are distinct from * each other * @see ReactiveX operators documentation: Distinct * @see #distinct(Function) * @see #distinct(Function, Callable) */ @SuppressWarnings({ "rawtypes", "unchecked" }) @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable distinct() { return distinct((Function)Functions.identity(), Functions.createHashSet()); } /** * Returns a Flowable that emits all items emitted by the source Publisher that are distinct according * to a key selector function and based on {@link Object#equals(Object)} comparison of the objects * returned by the key selector function. *

* *

* It is recommended the keys' class {@code K} overrides the default {@code Object.equals()} and {@link Object#hashCode()} to provide * a meaningful comparison between the key objects as the default Java implementation only considers reference equivalence. *

* By default, {@code distinct()} uses an internal {@link java.util.HashSet} per Subscriber to remember * previously seen keys and uses {@link java.util.Set#add(Object)} returning {@code false} as the * indicator for duplicates. *

* Note that this internal {@code HashSet} may grow unbounded as keys won't be removed from it by * the operator. Therefore, using very long or infinite upstream (with very distinct keys) may lead * to {@code OutOfMemoryError}. *

* Customizing the retention policy can happen only by providing a custom {@link java.util.Collection} implementation * to the {@link #distinct(Function, Callable)} overload. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code distinct} does not operate by default on a particular {@link Scheduler}.
*
* * @param the key type * @param keySelector * a function that projects an emitted item to a key value that is used to decide whether an item * is distinct from another one or not * @return a Flowable that emits those items emitted by the source Publisher that have distinct keys * @see ReactiveX operators documentation: Distinct * @see #distinct(Function, Callable) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable distinct(Function keySelector) { return distinct(keySelector, Functions.createHashSet()); } /** * Returns a Flowable that emits all items emitted by the source Publisher that are distinct according * to a key selector function and based on {@link Object#equals(Object)} comparison of the objects * returned by the key selector function. *

* *

* It is recommended the keys' class {@code K} overrides the default {@code Object.equals()} and {@link Object#hashCode()} to provide * a meaningful comparison between the key objects as the default Java implementation only considers reference equivalence. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code distinct} does not operate by default on a particular {@link Scheduler}.
*
* * @param the key type * @param keySelector * a function that projects an emitted item to a key value that is used to decide whether an item * is distinct from another one or not * @param collectionSupplier * function called for each individual Subscriber to return a Collection subtype for holding the extracted * keys and whose add() method's return indicates uniqueness. * @return a Flowable that emits those items emitted by the source Publisher that have distinct keys * @see ReactiveX operators documentation: Distinct */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable distinct(Function keySelector, Callable> collectionSupplier) { ObjectHelper.requireNonNull(keySelector, "keySelector is null"); ObjectHelper.requireNonNull(collectionSupplier, "collectionSupplier is null"); return RxJavaPlugins.onAssembly(new FlowableDistinct(this, keySelector, collectionSupplier)); } /** * Returns a Flowable that emits all items emitted by the source Publisher that are distinct from their * immediate predecessors based on {@link Object#equals(Object)} comparison. *

* *

* It is recommended the elements' class {@code T} in the flow overrides the default {@code Object.equals()} to provide * a meaningful comparison between items as the default Java implementation only considers reference equivalence. * Alternatively, use the {@link #distinctUntilChanged(BiPredicate)} overload and provide a comparison function * in case the class {@code T} can't be overridden with custom {@code equals()} or the comparison itself * should happen on different terms or properties of the class {@code T}. *

* Note that the operator always retains the latest item from upstream regardless of the comparison result * and uses it in the next comparison with the next upstream item. *

* Note that if element type {@code T} in the flow is mutable, the comparison of the previous and current * item may yield unexpected results if the items are mutated externally. Common cases are mutable * {@code CharSequence}s or {@code List}s where the objects will actually have the same * references when they are modified and {@code distinctUntilChanged} will evaluate subsequent items as same. * To avoid such situation, it is recommended that mutable data is converted to an immutable one, * for example using {@code map(CharSequence::toString)} or {@code map(list -> Collections.unmodifiableList(new ArrayList<>(list)))}. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code distinctUntilChanged} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Flowable that emits those items from the source Publisher that are distinct from their * immediate predecessors * @see ReactiveX operators documentation: Distinct * @see #distinctUntilChanged(BiPredicate) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable distinctUntilChanged() { return distinctUntilChanged(Functions.identity()); } /** * Returns a Flowable that emits all items emitted by the source Publisher that are distinct from their * immediate predecessors, according to a key selector function and based on {@link Object#equals(Object)} comparison * of those objects returned by the key selector function. *

* *

* It is recommended the keys' class {@code K} overrides the default {@code Object.equals()} to provide * a meaningful comparison between the key objects as the default Java implementation only considers reference equivalence. * Alternatively, use the {@link #distinctUntilChanged(BiPredicate)} overload and provide a comparison function * in case the class {@code K} can't be overridden with custom {@code equals()} or the comparison itself * should happen on different terms or properties of the item class {@code T} (for which the keys can be * derived via a similar selector). *

* Note that the operator always retains the latest key from upstream regardless of the comparison result * and uses it in the next comparison with the next key derived from the next upstream item. *

* Note that if element type {@code T} in the flow is mutable, the comparison of the previous and current * item may yield unexpected results if the items are mutated externally. Common cases are mutable * {@code CharSequence}s or {@code List}s where the objects will actually have the same * references when they are modified and {@code distinctUntilChanged} will evaluate subsequent items as same. * To avoid such situation, it is recommended that mutable data is converted to an immutable one, * for example using {@code map(CharSequence::toString)} or {@code map(list -> Collections.unmodifiableList(new ArrayList<>(list)))}. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code distinctUntilChanged} does not operate by default on a particular {@link Scheduler}.
*
* * @param the key type * @param keySelector * a function that projects an emitted item to a key value that is used to decide whether an item * is distinct from another one or not * @return a Flowable that emits those items from the source Publisher whose keys are distinct from * those of their immediate predecessors * @see ReactiveX operators documentation: Distinct */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable distinctUntilChanged(Function keySelector) { ObjectHelper.requireNonNull(keySelector, "keySelector is null"); return RxJavaPlugins.onAssembly(new FlowableDistinctUntilChanged(this, keySelector, ObjectHelper.equalsPredicate())); } /** * Returns a Flowable that emits all items emitted by the source Publisher that are distinct from their * immediate predecessors when compared with each other via the provided comparator function. *

* *

* Note that the operator always retains the latest item from upstream regardless of the comparison result * and uses it in the next comparison with the next upstream item. *

* Note that if element type {@code T} in the flow is mutable, the comparison of the previous and current * item may yield unexpected results if the items are mutated externally. Common cases are mutable * {@code CharSequence}s or {@code List}s where the objects will actually have the same * references when they are modified and {@code distinctUntilChanged} will evaluate subsequent items as same. * To avoid such situation, it is recommended that mutable data is converted to an immutable one, * for example using {@code map(CharSequence::toString)} or {@code map(list -> Collections.unmodifiableList(new ArrayList<>(list)))}. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code distinctUntilChanged} does not operate by default on a particular {@link Scheduler}.
*
* * @param comparer the function that receives the previous item and the current item and is * expected to return true if the two are equal, thus skipping the current value. * @return a Flowable that emits those items from the source Publisher that are distinct from their * immediate predecessors * @see ReactiveX operators documentation: Distinct * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable distinctUntilChanged(BiPredicate comparer) { ObjectHelper.requireNonNull(comparer, "comparer is null"); return RxJavaPlugins.onAssembly(new FlowableDistinctUntilChanged(this, Functions.identity(), comparer)); } /** * Calls the specified action after this Flowable signals onError or onCompleted or gets canceled by * the downstream. *

In case of a race between a terminal event and a cancellation, the provided {@code onFinally} action * is executed once per subscription. *

Note that the {@code onFinally} action is shared between subscriptions and as such * should be thread-safe. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code doFinally} does not operate by default on a particular {@link Scheduler}.
*
Operator-fusion:
*
This operator supports normal and conditional Subscribers as well as boundary-limited * synchronous or asynchronous queue-fusion.
*
*

History: 2.0.1 - experimental * @param onFinally the action called when this Flowable terminates or gets canceled * @return the new Flowable instance * @since 2.1 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doFinally(Action onFinally) { ObjectHelper.requireNonNull(onFinally, "onFinally is null"); return RxJavaPlugins.onAssembly(new FlowableDoFinally(this, onFinally)); } /** * Calls the specified consumer with the current item after this item has been emitted to the downstream. *

Note that the {@code onAfterNext} action is shared between subscriptions and as such * should be thread-safe. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code doAfterNext} does not operate by default on a particular {@link Scheduler}.
*
Operator-fusion:
*
This operator supports normal and conditional Subscribers as well as boundary-limited * synchronous or asynchronous queue-fusion.
*
*

History: 2.0.1 - experimental * @param onAfterNext the Consumer that will be called after emitting an item from upstream to the downstream * @return the new Flowable instance * @since 2.1 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doAfterNext(Consumer onAfterNext) { ObjectHelper.requireNonNull(onAfterNext, "onAfterNext is null"); return RxJavaPlugins.onAssembly(new FlowableDoAfterNext(this, onAfterNext)); } /** * Registers an {@link Action} to be called when this Publisher invokes either * {@link Subscriber#onComplete onComplete} or {@link Subscriber#onError onError}. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code doAfterTerminate} does not operate by default on a particular {@link Scheduler}.
*
* * @param onAfterTerminate * an {@link Action} to be invoked when the source Publisher finishes * @return a Flowable that emits the same items as the source Publisher, then invokes the * {@link Action} * @see ReactiveX operators documentation: Do * @see #doOnTerminate(Action) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doAfterTerminate(Action onAfterTerminate) { return doOnEach(Functions.emptyConsumer(), Functions.emptyConsumer(), Functions.EMPTY_ACTION, onAfterTerminate); } /** * Calls the cancel {@code Action} if the downstream cancels the sequence. *

* The action is shared between subscriptions and thus may be called concurrently from multiple * threads; the action must be thread-safe. *

* If the action throws a runtime exception, that exception is rethrown by the {@code onCancel()} call, * sometimes as a {@code CompositeException} if there were multiple exceptions along the way. *

* Note that terminal events trigger the action unless the {@code Publisher} is subscribed to via {@code unsafeSubscribe()}. *

* *

*
Backpressure:
*
{@code doOnCancel} does not interact with backpressure requests or value delivery; backpressure * behavior is preserved between its upstream and its downstream.
*
Scheduler:
*
{@code doOnCancel} does not operate by default on a particular {@link Scheduler}.
*
* * @param onCancel * the action that gets called when the source {@code Publisher}'s Subscription is canceled * @return the source {@code Publisher} modified so as to call this Action when appropriate * @see ReactiveX operators documentation: Do */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doOnCancel(Action onCancel) { return doOnLifecycle(Functions.emptyConsumer(), Functions.EMPTY_LONG_CONSUMER, onCancel); } /** * Modifies the source Publisher so that it invokes an action when it calls {@code onComplete}. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code doOnComplete} does not operate by default on a particular {@link Scheduler}.
*
* * @param onComplete * the action to invoke when the source Publisher calls {@code onComplete} * @return the source Publisher with the side-effecting behavior applied * @see ReactiveX operators documentation: Do */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doOnComplete(Action onComplete) { return doOnEach(Functions.emptyConsumer(), Functions.emptyConsumer(), onComplete, Functions.EMPTY_ACTION); } /** * Calls the appropriate onXXX consumer (shared between all subscribers) whenever a signal with the same type * passes through, before forwarding them to downstream. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code doOnEach} does not operate by default on a particular {@link Scheduler}.
*
* * @return the source Publisher with the side-effecting behavior applied * @see ReactiveX operators documentation: Do */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) private Flowable doOnEach(Consumer onNext, Consumer onError, Action onComplete, Action onAfterTerminate) { ObjectHelper.requireNonNull(onNext, "onNext is null"); ObjectHelper.requireNonNull(onError, "onError is null"); ObjectHelper.requireNonNull(onComplete, "onComplete is null"); ObjectHelper.requireNonNull(onAfterTerminate, "onAfterTerminate is null"); return RxJavaPlugins.onAssembly(new FlowableDoOnEach(this, onNext, onError, onComplete, onAfterTerminate)); } /** * Modifies the source Publisher so that it invokes an action for each item it emits. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code doOnEach} does not operate by default on a particular {@link Scheduler}.
*
* * @param onNotification * the action to invoke for each item emitted by the source Publisher * @return the source Publisher with the side-effecting behavior applied * @see ReactiveX operators documentation: Do */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doOnEach(final Consumer> onNotification) { ObjectHelper.requireNonNull(onNotification, "onNotification is null"); return doOnEach( Functions.notificationOnNext(onNotification), Functions.notificationOnError(onNotification), Functions.notificationOnComplete(onNotification), Functions.EMPTY_ACTION ); } /** * Modifies the source Publisher so that it notifies a Subscriber for each item and terminal event it emits. *

* In case the {@code onError} of the supplied Subscriber throws, the downstream will receive a composite * exception containing the original exception and the exception thrown by {@code onError}. If either the * {@code onNext} or the {@code onComplete} method of the supplied Subscriber throws, the downstream will be * terminated and will receive this thrown exception. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code doOnEach} does not operate by default on a particular {@link Scheduler}.
*
* * @param subscriber * the Subscriber to be notified about onNext, onError and onComplete events on its * respective methods before the actual downstream Subscriber gets notified. * @return the source Publisher with the side-effecting behavior applied * @see ReactiveX operators documentation: Do */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doOnEach(final Subscriber subscriber) { ObjectHelper.requireNonNull(subscriber, "subscriber is null"); return doOnEach( FlowableInternalHelper.subscriberOnNext(subscriber), FlowableInternalHelper.subscriberOnError(subscriber), FlowableInternalHelper.subscriberOnComplete(subscriber), Functions.EMPTY_ACTION); } /** * Modifies the source Publisher so that it invokes an action if it calls {@code onError}. *

* In case the {@code onError} action throws, the downstream will receive a composite exception containing * the original exception and the exception thrown by {@code onError}. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code doOnError} does not operate by default on a particular {@link Scheduler}.
*
* * @param onError * the action to invoke if the source Publisher calls {@code onError} * @return the source Publisher with the side-effecting behavior applied * @see ReactiveX operators documentation: Do */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doOnError(Consumer onError) { return doOnEach(Functions.emptyConsumer(), onError, Functions.EMPTY_ACTION, Functions.EMPTY_ACTION); } /** * Calls the appropriate onXXX method (shared between all Subscribers) for the lifecycle events of * the sequence (subscription, cancellation, requesting). *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code doOnLifecycle} does not operate by default on a particular {@link Scheduler}.
*
* * @param onSubscribe * a Consumer called with the Subscription sent via Subscriber.onSubscribe() * @param onRequest * a LongConsumer called with the request amount sent via Subscription.request() * @param onCancel * called when the downstream cancels the Subscription via cancel() * @return the source Publisher with the side-effecting behavior applied * @see ReactiveX operators documentation: Do */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doOnLifecycle(final Consumer onSubscribe, final LongConsumer onRequest, final Action onCancel) { ObjectHelper.requireNonNull(onSubscribe, "onSubscribe is null"); ObjectHelper.requireNonNull(onRequest, "onRequest is null"); ObjectHelper.requireNonNull(onCancel, "onCancel is null"); return RxJavaPlugins.onAssembly(new FlowableDoOnLifecycle(this, onSubscribe, onRequest, onCancel)); } /** * Modifies the source Publisher so that it invokes an action when it calls {@code onNext}. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code doOnNext} does not operate by default on a particular {@link Scheduler}.
*
* * @param onNext * the action to invoke when the source Publisher calls {@code onNext} * @return the source Publisher with the side-effecting behavior applied * @see ReactiveX operators documentation: Do */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doOnNext(Consumer onNext) { return doOnEach(onNext, Functions.emptyConsumer(), Functions.EMPTY_ACTION, Functions.EMPTY_ACTION); } /** * Modifies the source {@code Publisher} so that it invokes the given action when it receives a * request for more items. *

* Note: This operator is for tracing the internal behavior of back-pressure request * patterns and generally intended for debugging use. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code doOnRequest} does not operate by default on a particular {@link Scheduler}.
*
* * @param onRequest * the action that gets called when a Subscriber requests items from this * {@code Publisher} * @return the source {@code Publisher} modified so as to call this Action when appropriate * @see ReactiveX operators * documentation: Do * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doOnRequest(LongConsumer onRequest) { return doOnLifecycle(Functions.emptyConsumer(), onRequest, Functions.EMPTY_ACTION); } /** * Modifies the source {@code Publisher} so that it invokes the given action when it is subscribed from * its subscribers. Each subscription will result in an invocation of the given action except when the * source {@code Publisher} is reference counted, in which case the source {@code Publisher} will invoke * the given action for the first subscription. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code doOnSubscribe} does not operate by default on a particular {@link Scheduler}.
*
* * @param onSubscribe * the Consumer that gets called when a Subscriber subscribes to the current {@code Flowable} * @return the source {@code Publisher} modified so as to call this Consumer when appropriate * @see ReactiveX operators documentation: Do */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doOnSubscribe(Consumer onSubscribe) { return doOnLifecycle(onSubscribe, Functions.EMPTY_LONG_CONSUMER, Functions.EMPTY_ACTION); } /** * Modifies the source Publisher so that it invokes an action when it calls {@code onComplete} or * {@code onError}. *

* *

* This differs from {@code doAfterTerminate} in that this happens before the {@code onComplete} or * {@code onError} notification. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s * backpressure behavior.
*
Scheduler:
*
{@code doOnTerminate} does not operate by default on a particular {@link Scheduler}.
*
* * @param onTerminate * the action to invoke when the source Publisher calls {@code onComplete} or {@code onError} * @return the source Publisher with the side-effecting behavior applied * @see ReactiveX operators documentation: Do * @see #doAfterTerminate(Action) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable doOnTerminate(final Action onTerminate) { return doOnEach(Functions.emptyConsumer(), Functions.actionConsumer(onTerminate), onTerminate, Functions.EMPTY_ACTION); } /** * Returns a Maybe that emits the single item at a specified index in a sequence of emissions from * this Flowable or completes if this Flowable sequence has fewer elements than index. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code elementAt} does not operate by default on a particular {@link Scheduler}.
*
* * @param index * the zero-based index of the item to retrieve * @return a Maybe that emits a single item: the item at the specified position in the sequence of * those emitted by the source Publisher * @see ReactiveX operators documentation: ElementAt */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Maybe elementAt(long index) { if (index < 0) { throw new IndexOutOfBoundsException("index >= 0 required but it was " + index); } return RxJavaPlugins.onAssembly(new FlowableElementAtMaybe(this, index)); } /** * Returns a Single that emits the item found at a specified index in a sequence of emissions from * this Flowable, or a default item if that index is out of range. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code elementAt} does not operate by default on a particular {@link Scheduler}.
*
* * @param index * the zero-based index of the item to retrieve * @param defaultItem * the default item * @return a Single that emits the item at the specified position in the sequence emitted by the source * Publisher, or the default item if that index is outside the bounds of the source sequence * @throws IndexOutOfBoundsException * if {@code index} is less than 0 * @see ReactiveX operators documentation: ElementAt */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single elementAt(long index, T defaultItem) { if (index < 0) { throw new IndexOutOfBoundsException("index >= 0 required but it was " + index); } ObjectHelper.requireNonNull(defaultItem, "defaultItem is null"); return RxJavaPlugins.onAssembly(new FlowableElementAtSingle(this, index, defaultItem)); } /** * Returns a Single that emits the item found at a specified index in a sequence of emissions from * this Flowable or signals a {@link NoSuchElementException} if this Flowable has fewer elements than index. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded manner * (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code elementAtOrError} does not operate by default on a particular {@link Scheduler}.
*
* * @param index * the zero-based index of the item to retrieve * @return a Single that emits the item at the specified position in the sequence emitted by the source * Publisher, or the default item if that index is outside the bounds of the source sequence * @throws IndexOutOfBoundsException * if {@code index} is less than 0 * @see ReactiveX operators documentation: ElementAt */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single elementAtOrError(long index) { if (index < 0) { throw new IndexOutOfBoundsException("index >= 0 required but it was " + index); } return RxJavaPlugins.onAssembly(new FlowableElementAtSingle(this, index, null)); } /** * Filters items emitted by a Publisher by only emitting those that satisfy a specified predicate. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code filter} does not operate by default on a particular {@link Scheduler}.
*
* * @param predicate * a function that evaluates each item emitted by the source Publisher, returning {@code true} * if it passes the filter * @return a Flowable that emits only those items emitted by the source Publisher that the filter * evaluates as {@code true} * @see ReactiveX operators documentation: Filter */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable filter(Predicate predicate) { ObjectHelper.requireNonNull(predicate, "predicate is null"); return RxJavaPlugins.onAssembly(new FlowableFilter(this, predicate)); } /** * Returns a Maybe that emits only the very first item emitted by this Flowable or * completes if this Flowable is empty. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code firstElement} does not operate by default on a particular {@link Scheduler}.
*
* * @return the new Maybe instance * @see ReactiveX operators documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.SPECIAL) // take may trigger UNBOUNDED_IN @SchedulerSupport(SchedulerSupport.NONE) public final Maybe firstElement() { return elementAt(0); } /** * Returns a Single that emits only the very first item emitted by this Flowable, or a default * item if this Flowable completes without emitting anything. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code first} does not operate by default on a particular {@link Scheduler}.
*
* * @param defaultItem * the default item to emit if the source Publisher doesn't emit anything * @return a Single that emits only the very first item from the source, or a default item if the * source Publisher completes without emitting any items * @see ReactiveX operators documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.SPECIAL) // take may trigger UNBOUNDED_IN @SchedulerSupport(SchedulerSupport.NONE) public final Single first(T defaultItem) { return elementAt(0, defaultItem); } /** * Returns a Single that emits only the very first item emitted by this Flowable or * signals a {@link NoSuchElementException} if this Flowable is empty. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code firstOrError} does not operate by default on a particular {@link Scheduler}.
*
* * @return the new Single instance * @see ReactiveX operators documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.SPECIAL) // take may trigger UNBOUNDED_IN @SchedulerSupport(SchedulerSupport.NONE) public final Single firstOrError() { return elementAtOrError(0); } /** * Returns a Flowable that emits items based on applying a function that you supply to each item emitted * by the source Publisher, where that function returns a Publisher, and then merging those resulting * Publishers and emitting the results of this merger. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@link #bufferSize()} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the inner Publishers and the output type * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @return a Flowable that emits the result of applying the transformation function to each item emitted * by the source Publisher and merging the results of the Publishers obtained from this * transformation * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap(Function> mapper) { return flatMap(mapper, false, bufferSize(), bufferSize()); } /** * Returns a Flowable that emits items based on applying a function that you supply to each item emitted * by the source Publisher, where that function returns a Publisher, and then merging those resulting * Publishers and emitting the results of this merger. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@link #bufferSize()} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the inner Publishers and the output type * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @param delayErrors * if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate * if false, the first one signaling an exception will terminate the whole sequence immediately * @return a Flowable that emits the result of applying the transformation function to each item emitted * by the source Publisher and merging the results of the Publishers obtained from this * transformation * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap(Function> mapper, boolean delayErrors) { return flatMap(mapper, delayErrors, bufferSize(), bufferSize()); } /** * Returns a Flowable that emits items based on applying a function that you supply to each item emitted * by the source Publisher, where that function returns a Publisher, and then merging those resulting * Publishers and emitting the results of this merger, while limiting the maximum number of concurrent * subscriptions to these Publishers. * * *
*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the inner Publishers and the output type * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @return a Flowable that emits the result of applying the transformation function to each item emitted * by the source Publisher and merging the results of the Publishers obtained from this * transformation * @see ReactiveX operators documentation: FlatMap * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap(Function> mapper, int maxConcurrency) { return flatMap(mapper, false, maxConcurrency, bufferSize()); } /** * Returns a Flowable that emits items based on applying a function that you supply to each item emitted * by the source Publisher, where that function returns a Publisher, and then merging those resulting * Publishers and emitting the results of this merger, while limiting the maximum number of concurrent * subscriptions to these Publishers. * * *
*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the inner Publishers and the output type * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @param delayErrors * if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate * if false, the first one signaling an exception will terminate the whole sequence immediately * @return a Flowable that emits the result of applying the transformation function to each item emitted * by the source Publisher and merging the results of the Publishers obtained from this * transformation * @see ReactiveX operators documentation: FlatMap * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap(Function> mapper, boolean delayErrors, int maxConcurrency) { return flatMap(mapper, delayErrors, maxConcurrency, bufferSize()); } /** * Returns a Flowable that emits items based on applying a function that you supply to each item emitted * by the source Publisher, where that function returns a Publisher, and then merging those resulting * Publishers and emitting the results of this merger, while limiting the maximum number of concurrent * subscriptions to these Publishers. * * *
*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the inner Publishers and the output type * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @param delayErrors * if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate * if false, the first one signaling an exception will terminate the whole sequence immediately * @param bufferSize * the number of elements to prefetch from each inner Publisher * @return a Flowable that emits the result of applying the transformation function to each item emitted * by the source Publisher and merging the results of the Publishers obtained from this * transformation * @see ReactiveX operators documentation: FlatMap * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap(Function> mapper, boolean delayErrors, int maxConcurrency, int bufferSize) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); if (this instanceof ScalarCallable) { @SuppressWarnings("unchecked") T v = ((ScalarCallable)this).call(); if (v == null) { return empty(); } return FlowableScalarXMap.scalarXMap(v, mapper); } return RxJavaPlugins.onAssembly(new FlowableFlatMap(this, mapper, delayErrors, maxConcurrency, bufferSize)); } /** * Returns a Flowable that applies a function to each item emitted or notification raised by the source * Publisher and then flattens the Publishers returned from these functions and emits the resulting items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@link #bufferSize()} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the result type * @param onNextMapper * a function that returns a Publisher to merge for each item emitted by the source Publisher * @param onErrorMapper * a function that returns a Publisher to merge for an onError notification from the source * Publisher * @param onCompleteSupplier * a function that returns a Publisher to merge for an onComplete notification from the source * Publisher * @return a Flowable that emits the results of merging the Publishers returned from applying the * specified functions to the emissions and notifications of the source Publisher * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap( Function> onNextMapper, Function> onErrorMapper, Callable> onCompleteSupplier) { ObjectHelper.requireNonNull(onNextMapper, "onNextMapper is null"); ObjectHelper.requireNonNull(onErrorMapper, "onErrorMapper is null"); ObjectHelper.requireNonNull(onCompleteSupplier, "onCompleteSupplier is null"); return merge(new FlowableMapNotification>(this, onNextMapper, onErrorMapper, onCompleteSupplier)); } /** * Returns a Flowable that applies a function to each item emitted or notification raised by the source * Publisher and then flattens the Publishers returned from these functions and emits the resulting items, * while limiting the maximum number of concurrent subscriptions to these Publishers. * * *
*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the result type * @param onNextMapper * a function that returns a Publisher to merge for each item emitted by the source Publisher * @param onErrorMapper * a function that returns a Publisher to merge for an onError notification from the source * Publisher * @param onCompleteSupplier * a function that returns a Publisher to merge for an onComplete notification from the source * Publisher * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @return a Flowable that emits the results of merging the Publishers returned from applying the * specified functions to the emissions and notifications of the source Publisher * @see ReactiveX operators documentation: FlatMap * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap( Function> onNextMapper, Function> onErrorMapper, Callable> onCompleteSupplier, int maxConcurrency) { ObjectHelper.requireNonNull(onNextMapper, "onNextMapper is null"); ObjectHelper.requireNonNull(onErrorMapper, "onErrorMapper is null"); ObjectHelper.requireNonNull(onCompleteSupplier, "onCompleteSupplier is null"); return merge(new FlowableMapNotification>( this, onNextMapper, onErrorMapper, onCompleteSupplier), maxConcurrency); } /** * Returns a Flowable that emits the results of a specified function to the pair of values emitted by the * source Publisher and a specified collection Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the inner Publishers * @param * the type of items emitted by the combiner function * @param mapper * a function that returns a Publisher for each item emitted by the source Publisher * @param combiner * a function that combines one item emitted by each of the source and collection Publishers and * returns an item to be emitted by the resulting Publisher * @return a Flowable that emits the results of applying a function to a pair of values emitted by the * source Publisher and the collection Publisher * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap(Function> mapper, BiFunction combiner) { return flatMap(mapper, combiner, false, bufferSize(), bufferSize()); } /** * Returns a Flowable that emits the results of a specified function to the pair of values emitted by the * source Publisher and a specified collection Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@link #bufferSize()} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the inner Publishers * @param * the type of items emitted by the combiner functions * @param mapper * a function that returns a Publisher for each item emitted by the source Publisher * @param combiner * a function that combines one item emitted by each of the source and collection Publishers and * returns an item to be emitted by the resulting Publisher * @param delayErrors * if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate * if false, the first one signaling an exception will terminate the whole sequence immediately * @return a Flowable that emits the results of applying a function to a pair of values emitted by the * source Publisher and the collection Publisher * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap(Function> mapper, BiFunction combiner, boolean delayErrors) { return flatMap(mapper, combiner, delayErrors, bufferSize(), bufferSize()); } /** * Returns a Flowable that emits the results of a specified function to the pair of values emitted by the * source Publisher and a specified collection Publisher, while limiting the maximum number of concurrent * subscriptions to these Publishers. * * *
*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the inner Publishers * @param * the type of items emitted by the combiner function * @param mapper * a function that returns a Publisher for each item emitted by the source Publisher * @param combiner * a function that combines one item emitted by each of the source and collection Publishers and * returns an item to be emitted by the resulting Publisher * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @param delayErrors * if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate * if false, the first one signaling an exception will terminate the whole sequence immediately * @return a Flowable that emits the results of applying a function to a pair of values emitted by the * source Publisher and the collection Publisher * @see ReactiveX operators documentation: FlatMap * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap(Function> mapper, BiFunction combiner, boolean delayErrors, int maxConcurrency) { return flatMap(mapper, combiner, delayErrors, maxConcurrency, bufferSize()); } /** * Returns a Flowable that emits the results of a specified function to the pair of values emitted by the * source Publisher and a specified collection Publisher, while limiting the maximum number of concurrent * subscriptions to these Publishers. * * *
*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the inner Publishers * @param * the type of items emitted by the combiner function * @param mapper * a function that returns a Publisher for each item emitted by the source Publisher * @param combiner * a function that combines one item emitted by each of the source and collection Publishers and * returns an item to be emitted by the resulting Publisher * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @param delayErrors * if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate * if false, the first one signaling an exception will terminate the whole sequence immediately * @param bufferSize * the number of elements to prefetch from the inner Publishers. * @return a Flowable that emits the results of applying a function to a pair of values emitted by the * source Publisher and the collection Publisher * @see ReactiveX operators documentation: FlatMap * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap(final Function> mapper, final BiFunction combiner, boolean delayErrors, int maxConcurrency, int bufferSize) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.requireNonNull(combiner, "combiner is null"); ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return flatMap(FlowableInternalHelper.flatMapWithCombiner(mapper, combiner), delayErrors, maxConcurrency, bufferSize); } /** * Returns a Flowable that emits the results of a specified function to the pair of values emitted by the * source Publisher and a specified collection Publisher, while limiting the maximum number of concurrent * subscriptions to these Publishers. * * *
*
Backpressure:
*
The operator honors backpressure from downstream. The upstream Flowable is consumed * in a bounded manner (up to {@link #bufferSize()} outstanding request amount for items). * The inner {@code Publisher}s are expected to honor backpressure; if violated, * the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the inner Publishers * @param * the type of items emitted by the combiner function * @param mapper * a function that returns a Publisher for each item emitted by the source Publisher * @param combiner * a function that combines one item emitted by each of the source and collection Publishers and * returns an item to be emitted by the resulting Publisher * @param maxConcurrency * the maximum number of Publishers that may be subscribed to concurrently * @return a Flowable that emits the results of applying a function to a pair of values emitted by the * source Publisher and the collection Publisher * @see ReactiveX operators documentation: FlatMap * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMap(Function> mapper, BiFunction combiner, int maxConcurrency) { return flatMap(mapper, combiner, false, maxConcurrency, bufferSize()); } /** * Maps each element of the upstream Flowable into CompletableSources, subscribes to them and * waits until the upstream and all CompletableSources complete. *
*
Backpressure:
*
The operator consumes the upstream in an unbounded manner.
*
Scheduler:
*
{@code flatMapCompletable} does not operate by default on a particular {@link Scheduler}.
*
* @param mapper the function that received each source value and transforms them into CompletableSources. * @return the new Completable instance */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Completable flatMapCompletable(Function mapper) { return flatMapCompletable(mapper, false, Integer.MAX_VALUE); } /** * Maps each element of the upstream Flowable into CompletableSources, subscribes to them and * waits until the upstream and all CompletableSources complete, optionally delaying all errors. *
*
Backpressure:
*
If {@code maxConcurrency == Integer.MAX_VALUE} the operator consumes the upstream in an unbounded manner. * Otherwise, the operator expects the upstream to honor backpressure. If the upstream doesn't support backpressure * the operator behaves as if {@code maxConcurrency == Integer.MAX_VALUE} was used.
*
Scheduler:
*
{@code flatMapCompletable} does not operate by default on a particular {@link Scheduler}.
*
* @param mapper the function that received each source value and transforms them into CompletableSources. * @param delayErrors if true errors from the upstream and inner CompletableSources are delayed until each of them * terminates. * @param maxConcurrency the maximum number of active subscriptions to the CompletableSources. * @return the new Completable instance */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Completable flatMapCompletable(Function mapper, boolean delayErrors, int maxConcurrency) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency"); return RxJavaPlugins.onAssembly(new FlowableFlatMapCompletableCompletable(this, mapper, delayErrors, maxConcurrency)); } /** * Returns a Flowable that merges each item emitted by the source Publisher with the values in an * Iterable corresponding to that item that is generated by a selector. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s is * expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMapIterable} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of item emitted by the resulting Iterable * @param mapper * a function that returns an Iterable sequence of values for when given an item emitted by the * source Publisher * @return a Flowable that emits the results of merging the items emitted by the source Publisher with * the values in the Iterables corresponding to those items, as generated by {@code collectionSelector} * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMapIterable(final Function> mapper) { return flatMapIterable(mapper, bufferSize()); } /** * Returns a Flowable that merges each item emitted by the source Publisher with the values in an * Iterable corresponding to that item that is generated by a selector. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s is * expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMapIterable} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of item emitted by the resulting Iterable * @param mapper * a function that returns an Iterable sequence of values for when given an item emitted by the * source Publisher * @param bufferSize * the number of elements to prefetch from the current Flowable * @return a Flowable that emits the results of merging the items emitted by the source Publisher with * the values in the Iterables corresponding to those items, as generated by {@code collectionSelector} * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMapIterable(final Function> mapper, int bufferSize) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableFlattenIterable(this, mapper, bufferSize)); } /** * Returns a Flowable that emits the results of applying a function to the pair of values from the source * Publisher and an Iterable corresponding to that item that is generated by a selector. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and the source {@code Publisher}s is * consumed in an unbounded manner (i.e., no backpressure is applied to it).
*
Scheduler:
*
{@code flatMapIterable} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the collection element type * @param * the type of item emitted by the resulting Iterable * @param mapper * a function that returns an Iterable sequence of values for each item emitted by the source * Publisher * @param resultSelector * a function that returns an item based on the item emitted by the source Publisher and the * Iterable returned for that item by the {@code collectionSelector} * @return a Flowable that emits the items returned by {@code resultSelector} for each item in the source * Publisher * @see ReactiveX operators documentation: FlatMap */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMapIterable(final Function> mapper, final BiFunction resultSelector) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.requireNonNull(resultSelector, "resultSelector is null"); return flatMap(FlowableInternalHelper.flatMapIntoIterable(mapper), resultSelector, false, bufferSize(), bufferSize()); } /** * Returns a Flowable that merges each item emitted by the source Publisher with the values in an * Iterable corresponding to that item that is generated by a selector, while limiting the number of concurrent * subscriptions to these Publishers. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s is * expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will * signal a {@code MissingBackpressureException}.
*
Scheduler:
*
{@code flatMapIterable} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the element type of the inner Iterable sequences * @param * the type of item emitted by the resulting Publisher * @param mapper * a function that returns an Iterable sequence of values for when given an item emitted by the * source Publisher * @param resultSelector * a function that returns an item based on the item emitted by the source Publisher and the * Iterable returned for that item by the {@code collectionSelector} * @param prefetch * the number of elements to prefetch from the current Flowable * @return a Flowable that emits the results of merging the items emitted by the source Publisher with * the values in the Iterables corresponding to those items, as generated by {@code collectionSelector} * @see ReactiveX operators documentation: FlatMap * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMapIterable(final Function> mapper, final BiFunction resultSelector, int prefetch) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.requireNonNull(resultSelector, "resultSelector is null"); return flatMap(FlowableInternalHelper.flatMapIntoIterable(mapper), resultSelector, false, bufferSize(), prefetch); } /** * Maps each element of the upstream Flowable into MaybeSources, subscribes to all of them * and merges their onSuccess values, in no particular order, into a single Flowable sequence. *
*
Backpressure:
*
The operator consumes the upstream in an unbounded manner.
*
Scheduler:
*
{@code flatMapMaybe} does not operate by default on a particular {@link Scheduler}.
*
* @param the result value type * @param mapper the function that received each source value and transforms them into MaybeSources. * @return the new Flowable instance */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMapMaybe(Function> mapper) { return flatMapMaybe(mapper, false, Integer.MAX_VALUE); } /** * Maps each element of the upstream Flowable into MaybeSources, subscribes to at most * {@code maxConcurrency} MaybeSources at a time and merges their onSuccess values, * in no particular order, into a single Flowable sequence, optionally delaying all errors. *
*
Backpressure:
*
If {@code maxConcurrency == Integer.MAX_VALUE} the operator consumes the upstream in an unbounded manner. * Otherwise, the operator expects the upstream to honor backpressure. If the upstream doesn't support backpressure * the operator behaves as if {@code maxConcurrency == Integer.MAX_VALUE} was used.
*
Scheduler:
*
{@code flatMapMaybe} does not operate by default on a particular {@link Scheduler}.
*
* @param the result value type * @param mapper the function that received each source value and transforms them into MaybeSources. * @param delayErrors if true errors from the upstream and inner MaybeSources are delayed until each of them * terminates. * @param maxConcurrency the maximum number of active subscriptions to the MaybeSources. * @return the new Flowable instance */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMapMaybe(Function> mapper, boolean delayErrors, int maxConcurrency) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency"); return RxJavaPlugins.onAssembly(new FlowableFlatMapMaybe(this, mapper, delayErrors, maxConcurrency)); } /** * Maps each element of the upstream Flowable into SingleSources, subscribes to all of them * and merges their onSuccess values, in no particular order, into a single Flowable sequence. *
*
Backpressure:
*
The operator consumes the upstream in an unbounded manner.
*
Scheduler:
*
{@code flatMapSingle} does not operate by default on a particular {@link Scheduler}.
*
* @param the result value type * @param mapper the function that received each source value and transforms them into SingleSources. * @return the new Flowable instance */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMapSingle(Function> mapper) { return flatMapSingle(mapper, false, Integer.MAX_VALUE); } /** * Maps each element of the upstream Flowable into SingleSources, subscribes to at most * {@code maxConcurrency} SingleSources at a time and merges their onSuccess values, * in no particular order, into a single Flowable sequence, optionally delaying all errors. *
*
Backpressure:
*
If {@code maxConcurrency == Integer.MAX_VALUE} the operator consumes the upstream in an unbounded manner. * Otherwise, the operator expects the upstream to honor backpressure. If the upstream doesn't support backpressure * the operator behaves as if {@code maxConcurrency == Integer.MAX_VALUE} was used.
*
Scheduler:
*
{@code flatMapSingle} does not operate by default on a particular {@link Scheduler}.
*
* @param the result value type * @param mapper the function that received each source value and transforms them into SingleSources. * @param delayErrors if true errors from the upstream and inner SingleSources are delayed until each of them * terminates. * @param maxConcurrency the maximum number of active subscriptions to the SingleSources. * @return the new Flowable instance */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable flatMapSingle(Function> mapper, boolean delayErrors, int maxConcurrency) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency"); return RxJavaPlugins.onAssembly(new FlowableFlatMapSingle(this, mapper, delayErrors, maxConcurrency)); } /** * Subscribes to the {@link Publisher} and receives notifications for each element. *

* Alias to {@link #subscribe(Consumer)} *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no * backpressure is applied to it).
*
Scheduler:
*
{@code forEach} does not operate by default on a particular {@link Scheduler}.
*
* * @param onNext * {@link Consumer} to execute for each item. * @return * a Disposable that allows canceling an asynchronous sequence * @throws NullPointerException * if {@code onNext} is null * @see ReactiveX operators documentation: Subscribe */ @CheckReturnValue @BackpressureSupport(BackpressureKind.NONE) @SchedulerSupport(SchedulerSupport.NONE) public final Disposable forEach(Consumer onNext) { return subscribe(onNext); } /** * Subscribes to the {@link Publisher} and receives notifications for each element until the * onNext Predicate returns false. *

* If the Flowable emits an error, it is wrapped into an * {@link io.reactivex.exceptions.OnErrorNotImplementedException OnErrorNotImplementedException} * and routed to the RxJavaPlugins.onError handler. *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no * backpressure is applied to it).
*
Scheduler:
*
{@code forEachWhile} does not operate by default on a particular {@link Scheduler}.
*
* * @param onNext * {@link Predicate} to execute for each item. * @return * a {@link Disposable} that allows canceling an asynchronous sequence * @throws NullPointerException * if {@code onNext} is null * @see ReactiveX operators documentation: Subscribe */ @CheckReturnValue @BackpressureSupport(BackpressureKind.NONE) @SchedulerSupport(SchedulerSupport.NONE) public final Disposable forEachWhile(Predicate onNext) { return forEachWhile(onNext, Functions.ON_ERROR_MISSING, Functions.EMPTY_ACTION); } /** * Subscribes to the {@link Publisher} and receives notifications for each element and error events until the * onNext Predicate returns false. *
*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no * backpressure is applied to it).
*
Scheduler:
*
{@code forEachWhile} does not operate by default on a particular {@link Scheduler}.
*
* * @param onNext * {@link Predicate} to execute for each item. * @param onError * {@link Consumer} to execute when an error is emitted. * @return * a {@link Disposable} that allows canceling an asynchronous sequence * @throws NullPointerException * if {@code onNext} is null, or * if {@code onError} is null * @see ReactiveX operators documentation: Subscribe */ @CheckReturnValue @BackpressureSupport(BackpressureKind.NONE) @SchedulerSupport(SchedulerSupport.NONE) public final Disposable forEachWhile(Predicate onNext, Consumer onError) { return forEachWhile(onNext, onError, Functions.EMPTY_ACTION); } /** * Subscribes to the {@link Publisher} and receives notifications for each element and the terminal events until the * onNext Predicate returns false. *
*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no * backpressure is applied to it).
*
Scheduler:
*
{@code forEachWhile} does not operate by default on a particular {@link Scheduler}.
*
* * @param onNext * {@link Predicate} to execute for each item. * @param onError * {@link Consumer} to execute when an error is emitted. * @param onComplete * {@link Action} to execute when completion is signaled. * @return * a {@link Disposable} that allows canceling an asynchronous sequence * @throws NullPointerException * if {@code onNext} is null, or * if {@code onError} is null, or * if {@code onComplete} is null * @see ReactiveX operators documentation: Subscribe */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.NONE) @SchedulerSupport(SchedulerSupport.NONE) public final Disposable forEachWhile(final Predicate onNext, final Consumer onError, final Action onComplete) { ObjectHelper.requireNonNull(onNext, "onNext is null"); ObjectHelper.requireNonNull(onError, "onError is null"); ObjectHelper.requireNonNull(onComplete, "onComplete is null"); ForEachWhileSubscriber s = new ForEachWhileSubscriber(onNext, onError, onComplete); subscribe(s); return s; } /** * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedPublisher} allows only a single * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the * source terminates, the next emission by the source having the same key will trigger a new * {@code GroupedPublisher} emission. *

* *

* Note: A {@link GroupedFlowable} will cache the items it is to emit until such time as it * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those * {@code GroupedPublisher}s that do not concern you. Instead, you can signal to them that they may * discard their buffers by applying an operator like {@link #ignoreElements} to them. *

* Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise, * the unconsumed groups may starve other groups due to the internal backpressure * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency * value to be greater or equal to the expected number of groups, possibly using * {@code Integer.MAX_VALUE} if the number of expected groups is unknown. * *

*
Backpressure:
*
Both the returned and its inner {@code Publisher}s honor backpressure and the source {@code Publisher} * is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on * downstream consumption). Note that both the returned and its inner {@code Publisher}s use * unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that may * lead to {@code OutOfMemoryError}.
*
Scheduler:
*
{@code groupBy} does not operate by default on a particular {@link Scheduler}.
*
* * @param keySelector * a function that extracts the key for each item * @param * the key type * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a * unique key value and each of which emits those items from the source Publisher that share that * key value * @see ReactiveX operators documentation: GroupBy */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> groupBy(Function keySelector) { return groupBy(keySelector, Functions.identity(), false, bufferSize()); } /** * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedPublisher} allows only a single * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the * source terminates, the next emission by the source having the same key will trigger a new * {@code GroupedPublisher} emission. *

* *

* Note: A {@link GroupedFlowable} will cache the items it is to emit until such time as it * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those * {@code GroupedPublisher}s that do not concern you. Instead, you can signal to them that they may * discard their buffers by applying an operator like {@link #ignoreElements} to them. *

* Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise, * the unconsumed groups may starve other groups due to the internal backpressure * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency * value to be greater or equal to the expected number of groups, possibly using * {@code Integer.MAX_VALUE} if the number of expected groups is unknown. *

*
Backpressure:
*
Both the returned and its inner {@code Publisher}s honor backpressure and the source {@code Publisher} * is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on * downstream consumption). Note that both the returned and its inner {@code Publisher}s use * unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that may * lead to {@code OutOfMemoryError}.
*
Scheduler:
*
{@code groupBy} does not operate by default on a particular {@link Scheduler}.
*
* * @param keySelector * a function that extracts the key for each item * @param * the key type * @param delayError * if true, the exception from the current Flowable is delayed in each group until that specific group emitted * the normal values; if false, the exception bypasses values in the groups and is reported immediately. * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a * unique key value and each of which emits those items from the source Publisher that share that * key value * @see ReactiveX operators documentation: GroupBy */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> groupBy(Function keySelector, boolean delayError) { return groupBy(keySelector, Functions.identity(), delayError, bufferSize()); } /** * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedPublisher} allows only a single * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the * source terminates, the next emission by the source having the same key will trigger a new * {@code GroupedPublisher} emission. *

* *

* Note: A {@link GroupedFlowable} will cache the items it is to emit until such time as it * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those * {@code GroupedPublisher}s that do not concern you. Instead, you can signal to them that they may * discard their buffers by applying an operator like {@link #ignoreElements} to them. *

* Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise, * the unconsumed groups may starve other groups due to the internal backpressure * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency * value to be greater or equal to the expected number of groups, possibly using * {@code Integer.MAX_VALUE} if the number of expected groups is unknown. * *

*
Backpressure:
*
Both the returned and its inner {@code Publisher}s honor backpressure and the source {@code Publisher} * is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on * downstream consumption). Note that both the returned and its inner {@code Publisher}s use * unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that may * lead to {@code OutOfMemoryError}.
*
Scheduler:
*
{@code groupBy} does not operate by default on a particular {@link Scheduler}.
*
* * @param keySelector * a function that extracts the key for each item * @param valueSelector * a function that extracts the return element for each item * @param * the key type * @param * the element type * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a * unique key value and each of which emits those items from the source Publisher that share that * key value * @see ReactiveX operators documentation: GroupBy */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> groupBy(Function keySelector, Function valueSelector) { return groupBy(keySelector, valueSelector, false, bufferSize()); } /** * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedPublisher} allows only a single * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the * source terminates, the next emission by the source having the same key will trigger a new * {@code GroupedPublisher} emission. *

* *

* Note: A {@link GroupedFlowable} will cache the items it is to emit until such time as it * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those * {@code GroupedPublisher}s that do not concern you. Instead, you can signal to them that they may * discard their buffers by applying an operator like {@link #ignoreElements} to them. *

* Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise, * the unconsumed groups may starve other groups due to the internal backpressure * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency * value to be greater or equal to the expected number of groups, possibly using * {@code Integer.MAX_VALUE} if the number of expected groups is unknown. * *

*
Backpressure:
*
Both the returned and its inner {@code Publisher}s honor backpressure and the source {@code Publisher} * is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on * downstream consumption). Note that both the returned and its inner {@code Publisher}s use * unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that may * lead to {@code OutOfMemoryError}.
*
Scheduler:
*
{@code groupBy} does not operate by default on a particular {@link Scheduler}.
*
* * @param keySelector * a function that extracts the key for each item * @param valueSelector * a function that extracts the return element for each item * @param * the key type * @param * the element type * @param delayError * if true, the exception from the current Flowable is delayed in each group until that specific group emitted * the normal values; if false, the exception bypasses values in the groups and is reported immediately. * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a * unique key value and each of which emits those items from the source Publisher that share that * key value * @see ReactiveX operators documentation: GroupBy */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> groupBy(Function keySelector, Function valueSelector, boolean delayError) { return groupBy(keySelector, valueSelector, delayError, bufferSize()); } /** * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedPublisher} allows only a single * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the * source terminates, the next emission by the source having the same key will trigger a new * {@code GroupedPublisher} emission. *

* *

* Note: A {@link GroupedFlowable} will cache the items it is to emit until such time as it * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those * {@code GroupedPublisher}s that do not concern you. Instead, you can signal to them that they may * discard their buffers by applying an operator like {@link #ignoreElements} to them. *

* Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise, * the unconsumed groups may starve other groups due to the internal backpressure * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency * value to be greater or equal to the expected number of groups, possibly using * {@code Integer.MAX_VALUE} if the number of expected groups is unknown. * *

*
Backpressure:
*
Both the returned and its inner {@code Publisher}s honor backpressure and the source {@code Publisher} * is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on * downstream consumption). Note that both the returned and its inner {@code Publisher}s use * unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that may * lead to {@code OutOfMemoryError}.
*
Scheduler:
*
{@code groupBy} does not operate by default on a particular {@link Scheduler}.
*
* * @param keySelector * a function that extracts the key for each item * @param valueSelector * a function that extracts the return element for each item * @param delayError * if true, the exception from the current Flowable is delayed in each group until that specific group emitted * the normal values; if false, the exception bypasses values in the groups and is reported immediately. * @param bufferSize * the hint for how many {@link GroupedFlowable}s and element in each {@link GroupedFlowable} should be buffered * @param * the key type * @param * the element type * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a * unique key value and each of which emits those items from the source Publisher that share that * key value * @see ReactiveX operators documentation: GroupBy */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> groupBy(Function keySelector, Function valueSelector, boolean delayError, int bufferSize) { ObjectHelper.requireNonNull(keySelector, "keySelector is null"); ObjectHelper.requireNonNull(valueSelector, "valueSelector is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableGroupBy(this, keySelector, valueSelector, bufferSize, delayError, null)); } /** * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedFlowable} allows only a single * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the * source terminates, the next emission by the source having the same key will trigger a new * {@code GroupedPublisher} emission. The {@code evictingMapFactory} is used to create a map that will * be used to hold the {@link GroupedFlowable}s by key. The evicting map created by this factory must * notify the provided {@code Consumer} with the entry value (not the key!) when an entry in this * map has been evicted. The next source emission will bring about the completion of the evicted * {@link GroupedFlowable}s and the arrival of an item with the same key as a completed {@link GroupedFlowable} * will prompt the creation and emission of a new {@link GroupedFlowable} with that key. * *

A use case for specifying an {@code evictingMapFactory} is where the source is infinite and fast and * over time the number of keys grows enough to be a concern in terms of the memory footprint of the * internal hash map containing the {@link GroupedFlowable}s. * *

The map created by an {@code evictingMapFactory} must be thread-safe. * *

An example of an {@code evictingMapFactory} using CacheBuilder from the Guava library is below: * *


     * Function<Consumer<Object>, Map<Integer, Object>> evictingMapFactory =
     *   notify ->
     *       CacheBuilder
     *         .newBuilder()
     *         .maximumSize(3)
     *         .removalListener(entry -> {
     *              try {
     *                  // emit the value not the key!
     *                  notify.accept(entry.getValue());
     *              } catch (Exception e) {
     *                  throw new RuntimeException(e);
     *              }
     *            })
     *         .<Integer, Object> build()
     *         .asMap();
     *
     * // Emit 1000 items but ensure that the
     * // internal map never has more than 3 items in it
     * Flowable
     *   .range(1, 1000)
     *   // note that number of keys is 10
     *   .groupBy(x -> x % 10, x -> x, true, 16, evictingMapFactory)
     *   .flatMap(g -> g)
     *   .forEach(System.out::println);
     * 
* *

* *

* Note: A {@link GroupedFlowable} will cache the items it is to emit until such time as it * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those * {@code GroupedFlowable}s that do not concern you. Instead, you can signal to them that they may * discard their buffers by applying an operator like {@link #ignoreElements} to them. *

* Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise, * the unconsumed groups may starve other groups due to the internal backpressure * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency * value to be greater or equal to the expected number of groups, possibly using * {@code Integer.MAX_VALUE} if the number of expected groups is unknown. * *

*
Backpressure:
*
Both the returned and its inner {@code GroupedFlowable}s honor backpressure and the source {@code Publisher} * is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on * downstream consumption). Note that both the returned and its inner {@code GroupedFlowable}s use * unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that may * lead to {@code OutOfMemoryError}.
*
Scheduler:
*
{@code groupBy} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.10 - beta * @param keySelector * a function that extracts the key for each item * @param valueSelector * a function that extracts the return element for each item * @param delayError * if true, the exception from the current Flowable is delayed in each group until that specific group emitted * the normal values; if false, the exception bypasses values in the groups and is reported immediately. * @param bufferSize * the hint for how many {@link GroupedFlowable}s and element in each {@link GroupedFlowable} should be buffered * @param evictingMapFactory * The factory used to create a map that will be used by the implementation to hold the * {@link GroupedFlowable}s. The evicting map created by this factory must * notify the provided {@code Consumer} with the entry value (not the key!) when * an entry in this map has been evicted. The next source emission will bring about the * completion of the evicted {@link GroupedFlowable}s. See example above. * @param * the key type * @param * the element type * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a * unique key value and each of which emits those items from the source Publisher that share that * key value * @see ReactiveX operators documentation: GroupBy * * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> groupBy(Function keySelector, Function valueSelector, boolean delayError, int bufferSize, Function, ? extends Map> evictingMapFactory) { ObjectHelper.requireNonNull(keySelector, "keySelector is null"); ObjectHelper.requireNonNull(valueSelector, "valueSelector is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); ObjectHelper.requireNonNull(evictingMapFactory, "evictingMapFactory is null"); return RxJavaPlugins.onAssembly(new FlowableGroupBy(this, keySelector, valueSelector, bufferSize, delayError, evictingMapFactory)); } /** * Returns a Flowable that correlates two Publishers when they overlap in time and groups the results. *

* There are no guarantees in what order the items get combined when multiple * items from one or both source Publishers overlap. *

* *

*
Backpressure:
*
The operator doesn't support backpressure and consumes all participating {@code Publisher}s in * an unbounded mode (i.e., not applying any backpressure to them).
*
Scheduler:
*
{@code groupJoin} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the right Publisher source * @param the element type of the left duration Publishers * @param the element type of the right duration Publishers * @param the result type * @param other * the other Publisher to correlate items from the source Publisher with * @param leftEnd * a function that returns a Publisher whose emissions indicate the duration of the values of * the source Publisher * @param rightEnd * a function that returns a Publisher whose emissions indicate the duration of the values of * the {@code right} Publisher * @param resultSelector * a function that takes an item emitted by each Publisher and returns the value to be emitted * by the resulting Publisher * @return a Flowable that emits items based on combining those items emitted by the source Publishers * whose durations overlap * @see ReactiveX operators documentation: Join */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable groupJoin( Publisher other, Function> leftEnd, Function> rightEnd, BiFunction, ? extends R> resultSelector) { ObjectHelper.requireNonNull(other, "other is null"); ObjectHelper.requireNonNull(leftEnd, "leftEnd is null"); ObjectHelper.requireNonNull(rightEnd, "rightEnd is null"); ObjectHelper.requireNonNull(resultSelector, "resultSelector is null"); return RxJavaPlugins.onAssembly(new FlowableGroupJoin( this, other, leftEnd, rightEnd, resultSelector)); } /** * Hides the identity of this Flowable and its Subscription. *

Allows hiding extra features such as {@link Processor}'s * {@link Subscriber} methods or preventing certain identity-based * optimizations (fusion). *

*
Backpressure:
*
The operator is a pass-through for backpressure, the behavior is determined by the upstream's * backpressure behavior.
*
Scheduler:
*
{@code hide} does not operate by default on a particular {@link Scheduler}.
*
* @return the new Flowable instance * * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable hide() { return RxJavaPlugins.onAssembly(new FlowableHide(this)); } /** * Ignores all items emitted by the source Publisher and only calls {@code onComplete} or {@code onError}. *

* *

*
Backpressure:
*
This operator ignores backpressure as it doesn't emit any elements and consumes the source {@code Publisher} * in an unbounded manner (i.e., no backpressure is applied to it).
*
Scheduler:
*
{@code ignoreElements} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Completable that only calls {@code onComplete} or {@code onError}, based on which one is * called by the source Publisher * @see ReactiveX operators documentation: IgnoreElements */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Completable ignoreElements() { return RxJavaPlugins.onAssembly(new FlowableIgnoreElementsCompletable(this)); } /** * Returns a Single that emits {@code true} if the source Publisher is empty, otherwise {@code false}. *

* In Rx.Net this is negated as the {@code any} Subscriber but we renamed this in RxJava to better match Java * naming idioms. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code isEmpty} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Flowable that emits a Boolean * @see ReactiveX operators documentation: Contains */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single isEmpty() { return all(Functions.alwaysFalse()); } /** * Correlates the items emitted by two Publishers based on overlapping durations. *

* There are no guarantees in what order the items get combined when multiple * items from one or both source Publishers overlap. *

* *

*
Backpressure:
*
The operator doesn't support backpressure and consumes all participating {@code Publisher}s in * an unbounded mode (i.e., not applying any backpressure to them).
*
Scheduler:
*
{@code join} does not operate by default on a particular {@link Scheduler}.
*
* * @param the value type of the right Publisher source * @param the element type of the left duration Publishers * @param the element type of the right duration Publishers * @param the result type * @param other * the second Publisher to join items from * @param leftEnd * a function to select a duration for each item emitted by the source Publisher, used to * determine overlap * @param rightEnd * a function to select a duration for each item emitted by the {@code right} Publisher, used to * determine overlap * @param resultSelector * a function that computes an item to be emitted by the resulting Publisher for any two * overlapping items emitted by the two Publishers * @return a Flowable that emits items correlating to items emitted by the source Publishers that have * overlapping durations * @see ReactiveX operators documentation: Join */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable join( Publisher other, Function> leftEnd, Function> rightEnd, BiFunction resultSelector) { ObjectHelper.requireNonNull(other, "other is null"); ObjectHelper.requireNonNull(leftEnd, "leftEnd is null"); ObjectHelper.requireNonNull(rightEnd, "rightEnd is null"); ObjectHelper.requireNonNull(resultSelector, "resultSelector is null"); return RxJavaPlugins.onAssembly(new FlowableJoin( this, other, leftEnd, rightEnd, resultSelector)); } /** * Returns a Maybe that emits the last item emitted by this Flowable or completes if * this Flowable is empty. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code lastElement} does not operate by default on a particular {@link Scheduler}.
*
* * @return a new Maybe instance * @see ReactiveX operators documentation: Last */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Maybe lastElement() { return RxJavaPlugins.onAssembly(new FlowableLastMaybe(this)); } /** * Returns a Single that emits only the last item emitted by this Flowable, or a default item * if this Flowable completes without emitting any items. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code last} does not operate by default on a particular {@link Scheduler}.
*
* * @param defaultItem * the default item to emit if the source Publisher is empty * @return the new Single instance * @see ReactiveX operators documentation: Last */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single last(T defaultItem) { ObjectHelper.requireNonNull(defaultItem, "defaultItem"); return RxJavaPlugins.onAssembly(new FlowableLastSingle(this, defaultItem)); } /** * Returns a Single that emits only the last item emitted by this Flowable or signals * a {@link NoSuchElementException} if this Flowable is empty. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code lastOrError} does not operate by default on a particular {@link Scheduler}.
*
* * @return the new Single instance * @see ReactiveX operators documentation: Last */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single lastOrError() { return RxJavaPlugins.onAssembly(new FlowableLastSingle(this, null)); } /** * This method requires advanced knowledge about building operators, please consider * other standard composition methods first; * Returns a {@code Flowable} which, when subscribed to, invokes the {@link FlowableOperator#apply(Subscriber) apply(Subscriber)} method * of the provided {@link FlowableOperator} for each individual downstream {@link Subscriber} and allows the * insertion of a custom operator by accessing the downstream's {@link Subscriber} during this subscription phase * and providing a new {@code Subscriber}, containing the custom operator's intended business logic, that will be * used in the subscription process going further upstream. *

* Generally, such a new {@code Subscriber} will wrap the downstream's {@code Subscriber} and forwards the * {@code onNext}, {@code onError} and {@code onComplete} events from the upstream directly or according to the * emission pattern the custom operator's business logic requires. In addition, such operator can intercept the * flow control calls of {@code cancel} and {@code request} that would have traveled upstream and perform * additional actions depending on the same business logic requirements. *

* Example: *


     * // Step 1: Create the consumer type that will be returned by the FlowableOperator.apply():
     * 
     * public final class CustomSubscriber<T> implements FlowableSubscriber<T>, Subscription {
     *
     *     // The downstream's Subscriber that will receive the onXXX events
     *     final Subscriber<? super String> downstream;
     *
     *     // The connection to the upstream source that will call this class' onXXX methods
     *     Subscription upstream;
     *
     *     // The constructor takes the downstream subscriber and usually any other parameters
     *     public CustomSubscriber(Subscriber<? super String> downstream) {
     *         this.downstream = downstream;
     *     }
     *
     *     // In the subscription phase, the upstream sends a Subscription to this class
     *     // and subsequently this class has to send a Subscription to the downstream.
     *     // Note that relaying the upstream's Subscription instance directly is not allowed in RxJava
     *     @Override
     *     public void onSubscribe(Subscription s) {
     *         if (upstream != null) {
     *             s.cancel();
     *         } else {
     *             upstream = s;
     *             downstream.onSubscribe(this);
     *         }
     *     }
     *
     *     // The upstream calls this with the next item and the implementation's
     *     // responsibility is to emit an item to the downstream based on the intended
     *     // business logic, or if it can't do so for the particular item,
     *     // request more from the upstream
     *     @Override
     *     public void onNext(T item) {
     *         String str = item.toString();
     *         if (str.length() < 2) {
     *             downstream.onNext(str);
     *         } else {
     *             upstream.request(1);
     *         }
     *     }
     *
     *     // Some operators may handle the upstream's error while others
     *     // could just forward it to the downstream.
     *     @Override
     *     public void onError(Throwable throwable) {
     *         downstream.onError(throwable);
     *     }
     *
     *     // When the upstream completes, usually the downstream should complete as well.
     *     @Override
     *     public void onComplete() {
     *         downstream.onComplete();
     *     }
     *
     *     // Some operators have to intercept the downstream's request calls to trigger
     *     // the emission of queued items while others can simply forward the request
     *     // amount as is.
     *     @Override
     *     public void request(long n) {
     *         upstream.request(n);
     *     }
     *
     *     // Some operators may use their own resources which should be cleaned up if
     *     // the downstream cancels the flow before it completed. Operators without
     *     // resources can simply forward the cancellation to the upstream.
     *     // In some cases, a canceled flag may be set by this method so that other parts
     *     // of this class may detect the cancellation and stop sending events
     *     // to the downstream.
     *     @Override
     *     public void cancel() {
     *         upstream.cancel();
     *     }
     * }
     *
     * // Step 2: Create a class that implements the FlowableOperator interface and
     * //         returns the custom consumer type from above in its apply() method.
     * //         Such class may define additional parameters to be submitted to
     * //         the custom consumer type.
     *
     * final class CustomOperator<T> implements FlowableOperator<String> {
     *     @Override
     *     public Subscriber<? super String> apply(Subscriber<? super T> upstream) {
     *         return new CustomSubscriber<T>(upstream);
     *     }
     * }
     *
     * // Step 3: Apply the custom operator via lift() in a flow by creating an instance of it
     * //         or reusing an existing one.
     *
     * Flowable.range(5, 10)
     * .lift(new CustomOperator<Integer>())
     * .test()
     * .assertResult("5", "6", "7", "8", "9");
     * 
*

* Creating custom operators can be complicated and it is recommended one consults the * RxJava wiki: Writing operators page about * the tools, requirements, rules, considerations and pitfalls of implementing them. *

* Note that implementing custom operators via this {@code lift()} method adds slightly more overhead by requiring * an additional allocation and indirection per assembled flows. Instead, extending the abstract {@code Flowable} * class and creating a {@link FlowableTransformer} with it is recommended. *

* Note also that it is not possible to stop the subscription phase in {@code lift()} as the {@code apply()} method * requires a non-null {@code Subscriber} instance to be returned, which is then unconditionally subscribed to * the upstream {@code Flowable}. For example, if the operator decided there is no reason to subscribe to the * upstream source because of some optimization possibility or a failure to prepare the operator, it still has to * return a {@code Subscriber} that should immediately cancel the upstream's {@code Subscription} in its * {@code onSubscribe} method. Again, using a {@code FlowableTransformer} and extending the {@code Flowable} is * a better option as {@link #subscribeActual} can decide to not subscribe to its upstream after all. *

*
Backpressure:
*
The {@code Subscriber} instance returned by the {@link FlowableOperator} is responsible to be * backpressure-aware or document the fact that the consumer of the returned {@code Publisher} has to apply one of * the {@code onBackpressureXXX} operators.
*
Scheduler:
*
{@code lift} does not operate by default on a particular {@link Scheduler}, however, the * {@link FlowableOperator} may use a {@code Scheduler} to support its own asynchronous behavior.
*
* * @param the output value type * @param lifter the {@link FlowableOperator} that receives the downstream's {@code Subscriber} and should return * a {@code Subscriber} with custom behavior to be used as the consumer for the current * {@code Flowable}. * @return the new Flowable instance * @see RxJava wiki: Writing operators * @see #compose(FlowableTransformer) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable lift(FlowableOperator lifter) { ObjectHelper.requireNonNull(lifter, "lifter is null"); return RxJavaPlugins.onAssembly(new FlowableLift(this, lifter)); } /** * Limits both the number of upstream items (after which the sequence completes) * and the total downstream request amount requested from the upstream to * possibly prevent the creation of excess items by the upstream. *

* The operator requests at most the given {@code count} of items from upstream even * if the downstream requests more than that. For example, given a {@code limit(5)}, * if the downstream requests 1, a request of 1 is submitted to the upstream * and the operator remembers that only 4 items can be requested now on. A request * of 5 at this point will request 4 from the upstream and any subsequent requests will * be ignored. *

* Note that requests are negotiated on an operator boundary and {@code limit}'s amount * may not be preserved further upstream. For example, * {@code source.observeOn(Schedulers.computation()).limit(5)} will still request the * default (128) elements from the given {@code source}. *

* The main use of this operator is with sources that are async boundaries that * don't interfere with request amounts, such as certain {@code Flowable}-based * network endpoints that relay downstream request amounts unchanged and are, therefore, * prone to trigger excessive item creation/transmission over the network. *

*
Backpressure:
*
The operator requests a total of the given {@code count} items from the upstream.
*
Scheduler:
*
{@code limit} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.6 - experimental * @param count the maximum number of items and the total request amount, non-negative. * Zero will immediately cancel the upstream on subscription and complete * the downstream. * @return the new Flowable instance * @see #take(long) * @see #rebatchRequests(int) * @since 2.2 */ @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue public final Flowable limit(long count) { if (count < 0) { throw new IllegalArgumentException("count >= 0 required but it was " + count); } return RxJavaPlugins.onAssembly(new FlowableLimit(this, count)); } /** * Returns a Flowable that applies a specified function to each item emitted by the source Publisher and * emits the results of these function applications. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code map} does not operate by default on a particular {@link Scheduler}.
*
* * @param the output type * @param mapper * a function to apply to each item emitted by the Publisher * @return a Flowable that emits the items from the source Publisher, transformed by the specified * function * @see ReactiveX operators documentation: Map */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable map(Function mapper) { ObjectHelper.requireNonNull(mapper, "mapper is null"); return RxJavaPlugins.onAssembly(new FlowableMap(this, mapper)); } /** * Returns a Flowable that represents all of the emissions and notifications from the source * Publisher into emissions marked with their original types within {@link Notification} objects. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and expects it from the source {@code Publisher}. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code materialize} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Flowable that emits items that are the result of materializing the items and notifications * of the source Publisher * @see ReactiveX operators documentation: Materialize * @see #dematerialize(Function) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> materialize() { return RxJavaPlugins.onAssembly(new FlowableMaterialize(this)); } /** * Flattens this and another Publisher into a single Publisher, without any transformation. *

* *

* You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by * using the {@code mergeWith} method. *

*
Backpressure:
*
The operator honors backpressure from downstream. This and the other {@code Publisher}s are expected to honor * backpressure; if violated, the operator may signal {@code MissingBackpressureException}.
*
Scheduler:
*
{@code mergeWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param other * a Publisher to be merged * @return a Flowable that emits all of the items emitted by the source Publishers * @see ReactiveX operators documentation: Merge */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable mergeWith(Publisher other) { ObjectHelper.requireNonNull(other, "other is null"); return merge(this, other); } /** * Merges the sequence of items of this Flowable with the success value of the other SingleSource. *

* *

* The success value of the other {@code SingleSource} can get interleaved at any point of this * {@code Flowable} sequence. *

*
Backpressure:
*
The operator honors backpressure from downstream and ensures the success item from the * {@code SingleSource} is emitted only when there is a downstream demand.
*
Scheduler:
*
{@code mergeWith} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.10 - experimental * @param other the {@code SingleSource} whose success value to merge with * @return the new Flowable instance * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable mergeWith(@NonNull SingleSource other) { ObjectHelper.requireNonNull(other, "other is null"); return RxJavaPlugins.onAssembly(new FlowableMergeWithSingle(this, other)); } /** * Merges the sequence of items of this Flowable with the success value of the other MaybeSource * or waits for both to complete normally if the MaybeSource is empty. *

* *

* The success value of the other {@code MaybeSource} can get interleaved at any point of this * {@code Flowable} sequence. *

*
Backpressure:
*
The operator honors backpressure from downstream and ensures the success item from the * {@code MaybeSource} is emitted only when there is a downstream demand.
*
Scheduler:
*
{@code mergeWith} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.10 - experimental * @param other the {@code MaybeSource} which provides a success value to merge with or completes * @return the new Flowable instance * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable mergeWith(@NonNull MaybeSource other) { ObjectHelper.requireNonNull(other, "other is null"); return RxJavaPlugins.onAssembly(new FlowableMergeWithMaybe(this, other)); } /** * Relays the items of this Flowable and completes only when the other CompletableSource completes * as well. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code mergeWith} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.10 - experimental * @param other the {@code CompletableSource} to await for completion * @return the new Flowable instance * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable mergeWith(@NonNull CompletableSource other) { ObjectHelper.requireNonNull(other, "other is null"); return RxJavaPlugins.onAssembly(new FlowableMergeWithCompletable(this, other)); } /** * Modifies a Publisher to perform its emissions and notifications on a specified {@link Scheduler}, * asynchronously with a bounded buffer of {@link #bufferSize()} slots. * *

Note that onError notifications will cut ahead of onNext notifications on the emission thread if Scheduler is truly * asynchronous. If strict event ordering is required, consider using the {@link #observeOn(Scheduler, boolean)} overload. *

* *

* This operator keeps emitting as many signals as it can on the given Scheduler's Worker thread, * which may result in a longer than expected occupation of this thread. In other terms, * it does not allow per-signal fairness in case the worker runs on a shared underlying thread. * If such fairness and signal/work interleaving is preferred, use the delay operator with zero time instead. *

*
Backpressure:
*
This operator honors backpressure from downstream and expects it from the source {@code Publisher}. Violating this * expectation will lead to {@code MissingBackpressureException}. This is the most common operator where the exception * pops up; look for sources up the chain that don't support backpressure, * such as {@code interval}, {@code timer}, {code PublishSubject} or {@code BehaviorSubject} and apply any * of the {@code onBackpressureXXX} operators before applying {@code observeOn} itself.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param scheduler * the {@link Scheduler} to notify {@link Subscriber}s on * @return the source Publisher modified so that its {@link Subscriber}s are notified on the specified * {@link Scheduler} * @see ReactiveX operators documentation: ObserveOn * @see RxJava Threading Examples * @see #subscribeOn * @see #observeOn(Scheduler, boolean) * @see #observeOn(Scheduler, boolean, int) * @see #delay(long, TimeUnit, Scheduler) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable observeOn(Scheduler scheduler) { return observeOn(scheduler, false, bufferSize()); } /** * Modifies a Publisher to perform its emissions and notifications on a specified {@link Scheduler}, * asynchronously with a bounded buffer and optionally delays onError notifications. *

* *

* This operator keeps emitting as many signals as it can on the given Scheduler's Worker thread, * which may result in a longer than expected occupation of this thread. In other terms, * it does not allow per-signal fairness in case the worker runs on a shared underlying thread. * If such fairness and signal/work interleaving is preferred, use the delay operator with zero time instead. *

*
Backpressure:
*
This operator honors backpressure from downstream and expects it from the source {@code Publisher}. Violating this * expectation will lead to {@code MissingBackpressureException}. This is the most common operator where the exception * pops up; look for sources up the chain that don't support backpressure, * such as {@code interval}, {@code timer}, {code PublishSubject} or {@code BehaviorSubject} and apply any * of the {@code onBackpressureXXX} operators before applying {@code observeOn} itself.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param scheduler * the {@link Scheduler} to notify {@link Subscriber}s on * @param delayError * indicates if the onError notification may not cut ahead of onNext notification on the other side of the * scheduling boundary. If true a sequence ending in onError will be replayed in the same order as was received * from upstream * @return the source Publisher modified so that its {@link Subscriber}s are notified on the specified * {@link Scheduler} * @see ReactiveX operators documentation: ObserveOn * @see RxJava Threading Examples * @see #subscribeOn * @see #observeOn(Scheduler) * @see #observeOn(Scheduler, boolean, int) * @see #delay(long, TimeUnit, Scheduler, boolean) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable observeOn(Scheduler scheduler, boolean delayError) { return observeOn(scheduler, delayError, bufferSize()); } /** * Modifies a Publisher to perform its emissions and notifications on a specified {@link Scheduler}, * asynchronously with a bounded buffer of configurable size and optionally delays onError notifications. *

* *

* This operator keeps emitting as many signals as it can on the given Scheduler's Worker thread, * which may result in a longer than expected occupation of this thread. In other terms, * it does not allow per-signal fairness in case the worker runs on a shared underlying thread. * If such fairness and signal/work interleaving is preferred, use the delay operator with zero time instead. *

*
Backpressure:
*
This operator honors backpressure from downstream and expects it from the source {@code Publisher}. Violating this * expectation will lead to {@code MissingBackpressureException}. This is the most common operator where the exception * pops up; look for sources up the chain that don't support backpressure, * such as {@code interval}, {@code timer}, {code PublishSubject} or {@code BehaviorSubject} and apply any * of the {@code onBackpressureXXX} operators before applying {@code observeOn} itself.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param scheduler * the {@link Scheduler} to notify {@link Subscriber}s on * @param delayError * indicates if the onError notification may not cut ahead of onNext notification on the other side of the * scheduling boundary. If true a sequence ending in onError will be replayed in the same order as was received * from upstream * @param bufferSize the size of the buffer. * @return the source Publisher modified so that its {@link Subscriber}s are notified on the specified * {@link Scheduler} * @see ReactiveX operators documentation: ObserveOn * @see RxJava Threading Examples * @see #subscribeOn * @see #observeOn(Scheduler) * @see #observeOn(Scheduler, boolean) * @see #delay(long, TimeUnit, Scheduler, boolean) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable observeOn(Scheduler scheduler, boolean delayError, int bufferSize) { ObjectHelper.requireNonNull(scheduler, "scheduler is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableObserveOn(this, scheduler, delayError, bufferSize)); } /** * Filters the items emitted by a Publisher, only emitting those of the specified type. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code ofType} does not operate by default on a particular {@link Scheduler}.
*
* * @param the output type * @param clazz * the class type to filter the items emitted by the source Publisher * @return a Flowable that emits items from the source Publisher of type {@code clazz} * @see ReactiveX operators documentation: Filter */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable ofType(final Class clazz) { ObjectHelper.requireNonNull(clazz, "clazz is null"); return filter(Functions.isInstanceOf(clazz)).cast(clazz); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer these * items indefinitely until they can be emitted. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.
*
* * @return the source Publisher modified to buffer items to the extent system resources allow * @see ReactiveX operators documentation: backpressure operators */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureBuffer() { return onBackpressureBuffer(bufferSize(), false, true); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer these * items indefinitely until they can be emitted. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.
*
* @param delayError * if true, an exception from the current Flowable is delayed until all buffered elements have been * consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping * any buffered element * @return the source Publisher modified to buffer items to the extent system resources allow * @see ReactiveX operators documentation: backpressure operators */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureBuffer(boolean delayError) { return onBackpressureBuffer(bufferSize(), delayError, true); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to * a given amount of items until they can be emitted. The resulting Publisher will signal * a {@code BufferOverflowException} via {@code onError} as soon as the buffer's capacity is exceeded, dropping all undelivered * items, and canceling the source. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param capacity number of slots available in the buffer. * @return the source {@code Publisher} modified to buffer items up to the given capacity. * @see ReactiveX operators documentation: backpressure operators * @since 1.1.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureBuffer(int capacity) { return onBackpressureBuffer(capacity, false, false); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to * a given amount of items until they can be emitted. The resulting Publisher will signal * a {@code BufferOverflowException} via {@code onError} as soon as the buffer's capacity is exceeded, dropping all undelivered * items, and canceling the source. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param capacity number of slots available in the buffer. * @param delayError * if true, an exception from the current Flowable is delayed until all buffered elements have been * consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping * any buffered element * @return the source {@code Publisher} modified to buffer items up to the given capacity. * @see ReactiveX operators documentation: backpressure operators * @since 1.1.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureBuffer(int capacity, boolean delayError) { return onBackpressureBuffer(capacity, delayError, false); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to * a given amount of items until they can be emitted. The resulting Publisher will signal * a {@code BufferOverflowException} via {@code onError} as soon as the buffer's capacity is exceeded, dropping all undelivered * items, and canceling the source. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param capacity number of slots available in the buffer. * @param delayError * if true, an exception from the current Flowable is delayed until all buffered elements have been * consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping * any buffered element * @param unbounded * if true, the capacity value is interpreted as the internal "island" size of the unbounded buffer * @return the source {@code Publisher} modified to buffer items up to the given capacity. * @see ReactiveX operators documentation: backpressure operators * @since 1.1.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureBuffer(int capacity, boolean delayError, boolean unbounded) { ObjectHelper.verifyPositive(capacity, "capacity"); return RxJavaPlugins.onAssembly(new FlowableOnBackpressureBuffer(this, capacity, unbounded, delayError, Functions.EMPTY_ACTION)); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to * a given amount of items until they can be emitted. The resulting Publisher will signal * a {@code BufferOverflowException} via {@code onError} as soon as the buffer's capacity is exceeded, dropping all undelivered * items, canceling the source, and notifying the producer with {@code onOverflow}. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param capacity number of slots available in the buffer. * @param delayError * if true, an exception from the current Flowable is delayed until all buffered elements have been * consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping * any buffered element * @param unbounded * if true, the capacity value is interpreted as the internal "island" size of the unbounded buffer * @param onOverflow action to execute if an item needs to be buffered, but there are no available slots. Null is allowed. * @return the source {@code Publisher} modified to buffer items up to the given capacity * @see ReactiveX operators documentation: backpressure operators * @since 1.1.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureBuffer(int capacity, boolean delayError, boolean unbounded, Action onOverflow) { ObjectHelper.requireNonNull(onOverflow, "onOverflow is null"); ObjectHelper.verifyPositive(capacity, "capacity"); return RxJavaPlugins.onAssembly(new FlowableOnBackpressureBuffer(this, capacity, unbounded, delayError, onOverflow)); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to * a given amount of items until they can be emitted. The resulting Publisher will signal * a {@code BufferOverflowException} via {@code onError} as soon as the buffer's capacity is exceeded, dropping all undelivered * items, canceling the source, and notifying the producer with {@code onOverflow}. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param capacity number of slots available in the buffer. * @param onOverflow action to execute if an item needs to be buffered, but there are no available slots. Null is allowed. * @return the source {@code Publisher} modified to buffer items up to the given capacity * @see ReactiveX operators documentation: backpressure operators * @since 1.1.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureBuffer(int capacity, Action onOverflow) { return onBackpressureBuffer(capacity, false, false, onOverflow); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to * a given amount of items until they can be emitted. The resulting Publisher will behave as determined * by {@code overflowStrategy} if the buffer capacity is exceeded. * *
    *
  • {@code BackpressureOverflow.Strategy.ON_OVERFLOW_ERROR} (default) will call {@code onError} dropping all undelivered items, * canceling the source, and notifying the producer with {@code onOverflow}.
  • *
  • {@code BackpressureOverflow.Strategy.ON_OVERFLOW_DROP_LATEST} will drop any new items emitted by the producer while * the buffer is full, without generating any {@code onError}. Each drop will, however, invoke {@code onOverflow} * to signal the overflow to the producer.
  • *
  • {@code BackpressureOverflow.Strategy.ON_OVERFLOW_DROP_OLDEST} will drop the oldest items in the buffer in order to make * room for newly emitted ones. Overflow will not generate an{@code onError}, but each drop will invoke * {@code onOverflow} to signal the overflow to the producer.
  • *
* *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.
*
* * @param capacity number of slots available in the buffer. * @param onOverflow action to execute if an item needs to be buffered, but there are no available slots. Null is allowed. * @param overflowStrategy how should the {@code Publisher} react to buffer overflows. Null is not allowed. * @return the source {@code Flowable} modified to buffer items up to the given capacity * @see ReactiveX operators documentation: backpressure operators * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureBuffer(long capacity, Action onOverflow, BackpressureOverflowStrategy overflowStrategy) { ObjectHelper.requireNonNull(overflowStrategy, "overflowStrategy is null"); ObjectHelper.verifyPositive(capacity, "capacity"); return RxJavaPlugins.onAssembly(new FlowableOnBackpressureBufferStrategy(this, capacity, onOverflow, overflowStrategy)); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to discard, * rather than emit, those items that its Subscriber is not prepared to observe. *

* *

* If the downstream request count hits 0 then the Publisher will refrain from calling {@code onNext} until * the Subscriber invokes {@code request(n)} again to increase the request count. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureDrop} does not operate by default on a particular {@link Scheduler}.
*
* * @return the source Publisher modified to drop {@code onNext} notifications on overflow * @see ReactiveX operators documentation: backpressure operators */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureDrop() { return RxJavaPlugins.onAssembly(new FlowableOnBackpressureDrop(this)); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to discard, * rather than emit, those items that its Subscriber is not prepared to observe. *

* *

* If the downstream request count hits 0 then the Publisher will refrain from calling {@code onNext} until * the Subscriber invokes {@code request(n)} again to increase the request count. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureDrop} does not operate by default on a particular {@link Scheduler}.
*
* * @param onDrop the action to invoke for each item dropped. onDrop action should be fast and should never block. * @return the source Publisher modified to drop {@code onNext} notifications on overflow * @see ReactiveX operators documentation: backpressure operators * @since 1.1.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureDrop(Consumer onDrop) { ObjectHelper.requireNonNull(onDrop, "onDrop is null"); return RxJavaPlugins.onAssembly(new FlowableOnBackpressureDrop(this, onDrop)); } /** * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to * hold onto the latest value and emit that on request. *

* *

* Its behavior is logically equivalent to {@code blockingLatest()} with the exception that * the downstream is not blocking while requesting more values. *

* Note that if the upstream Publisher does support backpressure, this operator ignores that capability * and doesn't propagate any backpressure requests from downstream. *

* Note that due to the nature of how backpressure requests are propagated through subscribeOn/observeOn, * requesting more than 1 from downstream doesn't guarantee a continuous delivery of onNext events. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded * manner (i.e., not applying backpressure to it).
*
Scheduler:
*
{@code onBackpressureLatest} does not operate by default on a particular {@link Scheduler}.
*
* * @return the source Publisher modified so that it emits the most recently-received item upon request * @since 1.1.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onBackpressureLatest() { return RxJavaPlugins.onAssembly(new FlowableOnBackpressureLatest(this)); } /** * Instructs a Publisher to pass control to another Publisher rather than invoking * {@link Subscriber#onError onError} if it encounters an error. *

* *

* By default, when a Publisher encounters an error that prevents it from emitting the expected item to * its {@link Subscriber}, the Publisher invokes its Subscriber's {@code onError} method, and then quits * without invoking any more of its Subscriber's methods. The {@code onErrorResumeNext} method changes this * behavior. If you pass a function that returns a Publisher ({@code resumeFunction}) to * {@code onErrorResumeNext}, if the original Publisher encounters an error, instead of invoking its * Subscriber's {@code onError} method, it will instead relinquish control to the Publisher returned from * {@code resumeFunction}, which will invoke the Subscriber's {@link Subscriber#onNext onNext} method if it is * able to do so. In such a case, because no Publisher necessarily invokes {@code onError}, the Subscriber * may never know that an error happened. *

* You can use this to prevent errors from propagating or to supply fallback data should errors be * encountered. *

*
Backpressure:
*
The operator honors backpressure from downstream. This and the resuming {@code Publisher}s * are expected to honor backpressure as well. * If any of them violate this expectation, the operator may throw an * {@code IllegalStateException} when the source {@code Publisher} completes or * a {@code MissingBackpressureException} is signaled somewhere downstream.
*
Scheduler:
*
{@code onErrorResumeNext} does not operate by default on a particular {@link Scheduler}.
*
* * @param resumeFunction * a function that returns a Publisher that will take over if the source Publisher encounters * an error * @return the original Publisher, with appropriately modified behavior * @see ReactiveX operators documentation: Catch */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onErrorResumeNext(Function> resumeFunction) { ObjectHelper.requireNonNull(resumeFunction, "resumeFunction is null"); return RxJavaPlugins.onAssembly(new FlowableOnErrorNext(this, resumeFunction, false)); } /** * Instructs a Publisher to pass control to another Publisher rather than invoking * {@link Subscriber#onError onError} if it encounters an error. *

* *

* By default, when a Publisher encounters an error that prevents it from emitting the expected item to * its {@link Subscriber}, the Publisher invokes its Subscriber's {@code onError} method, and then quits * without invoking any more of its Subscriber's methods. The {@code onErrorResumeNext} method changes this * behavior. If you pass another Publisher ({@code resumeSequence}) to a Publisher's * {@code onErrorResumeNext} method, if the original Publisher encounters an error, instead of invoking its * Subscriber's {@code onError} method, it will instead relinquish control to {@code resumeSequence} which * will invoke the Subscriber's {@link Subscriber#onNext onNext} method if it is able to do so. In such a case, * because no Publisher necessarily invokes {@code onError}, the Subscriber may never know that an error * happened. *

* You can use this to prevent errors from propagating or to supply fallback data should errors be * encountered. *

*
Backpressure:
*
The operator honors backpressure from downstream. This and the resuming {@code Publisher}s * are expected to honor backpressure as well. * If any of them violate this expectation, the operator may throw an * {@code IllegalStateException} when the source {@code Publisher} completes or * {@code MissingBackpressureException} is signaled somewhere downstream.
*
Scheduler:
*
{@code onErrorResumeNext} does not operate by default on a particular {@link Scheduler}.
*
* * @param next * the next Publisher source that will take over if the source Publisher encounters * an error * @return the original Publisher, with appropriately modified behavior * @see ReactiveX operators documentation: Catch */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onErrorResumeNext(final Publisher next) { ObjectHelper.requireNonNull(next, "next is null"); return onErrorResumeNext(Functions.justFunction(next)); } /** * Instructs a Publisher to emit an item (returned by a specified function) rather than invoking * {@link Subscriber#onError onError} if it encounters an error. *

* *

* By default, when a Publisher encounters an error that prevents it from emitting the expected item to * its {@link Subscriber}, the Publisher invokes its Subscriber's {@code onError} method, and then quits * without invoking any more of its Subscriber's methods. The {@code onErrorReturn} method changes this * behavior. If you pass a function ({@code resumeFunction}) to a Publisher's {@code onErrorReturn} * method, if the original Publisher encounters an error, instead of invoking its Subscriber's * {@code onError} method, it will instead emit the return value of {@code resumeFunction}. *

* You can use this to prevent errors from propagating or to supply fallback data should errors be * encountered. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s is expected to honor * backpressure as well. If it this expectation is violated, the operator may throw * {@code IllegalStateException} when the source {@code Publisher} completes or * {@code MissingBackpressureException} is signaled somewhere downstream.
*
Scheduler:
*
{@code onErrorReturn} does not operate by default on a particular {@link Scheduler}.
*
* * @param valueSupplier * a function that returns a single value that will be emitted along with a regular onComplete in case * the current Flowable signals an onError event * @return the original Publisher with appropriately modified behavior * @see ReactiveX operators documentation: Catch */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onErrorReturn(Function valueSupplier) { ObjectHelper.requireNonNull(valueSupplier, "valueSupplier is null"); return RxJavaPlugins.onAssembly(new FlowableOnErrorReturn(this, valueSupplier)); } /** * Instructs a Publisher to emit an item (returned by a specified function) rather than invoking * {@link Subscriber#onError onError} if it encounters an error. *

* *

* By default, when a Publisher encounters an error that prevents it from emitting the expected item to * its {@link Subscriber}, the Publisher invokes its Subscriber's {@code onError} method, and then quits * without invoking any more of its Subscriber's methods. The {@code onErrorReturn} method changes this * behavior. If you pass a function ({@code resumeFunction}) to a Publisher's {@code onErrorReturn} * method, if the original Publisher encounters an error, instead of invoking its Subscriber's * {@code onError} method, it will instead emit the return value of {@code resumeFunction}. *

* You can use this to prevent errors from propagating or to supply fallback data should errors be * encountered. *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher}s is expected to honor * backpressure as well. If it this expectation is violated, the operator may throw * {@code IllegalStateException} when the source {@code Publisher} completes or * {@code MissingBackpressureException} is signaled somewhere downstream.
*
Scheduler:
*
{@code onErrorReturnItem} does not operate by default on a particular {@link Scheduler}.
*
* * @param item * the value that is emitted along with a regular onComplete in case the current * Flowable signals an exception * @return the original Publisher with appropriately modified behavior * @see ReactiveX operators documentation: Catch */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onErrorReturnItem(final T item) { ObjectHelper.requireNonNull(item, "item is null"); return onErrorReturn(Functions.justFunction(item)); } /** * Instructs a Publisher to pass control to another Publisher rather than invoking * {@link Subscriber#onError onError} if it encounters an {@link java.lang.Exception}. *

* This differs from {@link #onErrorResumeNext} in that this one does not handle {@link java.lang.Throwable} * or {@link java.lang.Error} but lets those continue through. *

* *

* By default, when a Publisher encounters an exception that prevents it from emitting the expected item * to its {@link Subscriber}, the Publisher invokes its Subscriber's {@code onError} method, and then quits * without invoking any more of its Subscriber's methods. The {@code onExceptionResumeNext} method changes * this behavior. If you pass another Publisher ({@code resumeSequence}) to a Publisher's * {@code onExceptionResumeNext} method, if the original Publisher encounters an exception, instead of * invoking its Subscriber's {@code onError} method, it will instead relinquish control to * {@code resumeSequence} which will invoke the Subscriber's {@link Subscriber#onNext onNext} method if it is * able to do so. In such a case, because no Publisher necessarily invokes {@code onError}, the Subscriber * may never know that an exception happened. *

* You can use this to prevent exceptions from propagating or to supply fallback data should exceptions be * encountered. *

*
Backpressure:
*
The operator honors backpressure from downstream. This and the resuming {@code Publisher}s * are expected to honor backpressure as well. * If any of them violate this expectation, the operator may throw an * {@code IllegalStateException} when the source {@code Publisher} completes or * {@code MissingBackpressureException} is signaled somewhere downstream.
*
Scheduler:
*
{@code onExceptionResumeNext} does not operate by default on a particular {@link Scheduler}.
*
* * @param next * the next Publisher that will take over if the source Publisher encounters * an exception * @return the original Publisher, with appropriately modified behavior * @see ReactiveX operators documentation: Catch */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onExceptionResumeNext(final Publisher next) { ObjectHelper.requireNonNull(next, "next is null"); return RxJavaPlugins.onAssembly(new FlowableOnErrorNext(this, Functions.justFunction(next), true)); } /** * Nulls out references to the upstream producer and downstream Subscriber if * the sequence is terminated or downstream cancels. *
*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code onTerminateDetach} does not operate by default on a particular {@link Scheduler}.
*
* @return a Flowable which nulls out references to the upstream producer and downstream Subscriber if * the sequence is terminated or downstream cancels * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable onTerminateDetach() { return RxJavaPlugins.onAssembly(new FlowableDetach(this)); } /** * Parallelizes the flow by creating multiple 'rails' (equal to the number of CPUs) * and dispatches the upstream items to them in a round-robin fashion. *

* Note that the rails don't execute in parallel on their own and one needs to * apply {@link ParallelFlowable#runOn(Scheduler)} to specify the Scheduler where * each rail will execute. *

* To merge the parallel 'rails' back into a single sequence, use {@link ParallelFlowable#sequential()}. *

* *

*
Backpressure:
*
The operator requires the upstream to honor backpressure and each 'rail' honors backpressure * as well.
*
Scheduler:
*
{@code parallel} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.0.5 - experimental; 2.1 - beta * @return the new ParallelFlowable instance * @since 2.2 */ @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue public final ParallelFlowable parallel() { return ParallelFlowable.from(this); } /** * Parallelizes the flow by creating the specified number of 'rails' * and dispatches the upstream items to them in a round-robin fashion. *

* Note that the rails don't execute in parallel on their own and one needs to * apply {@link ParallelFlowable#runOn(Scheduler)} to specify the Scheduler where * each rail will execute. *

* To merge the parallel 'rails' back into a single sequence, use {@link ParallelFlowable#sequential()}. *

* *

*
Backpressure:
*
The operator requires the upstream to honor backpressure and each 'rail' honors backpressure * as well.
*
Scheduler:
*
{@code parallel} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.0.5 - experimental; 2.1 - beta * @param parallelism the number of 'rails' to use * @return the new ParallelFlowable instance * @since 2.2 */ @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue public final ParallelFlowable parallel(int parallelism) { ObjectHelper.verifyPositive(parallelism, "parallelism"); return ParallelFlowable.from(this, parallelism); } /** * Parallelizes the flow by creating the specified number of 'rails' * and dispatches the upstream items to them in a round-robin fashion and * uses the defined per-'rail' prefetch amount. *

* Note that the rails don't execute in parallel on their own and one needs to * apply {@link ParallelFlowable#runOn(Scheduler)} to specify the Scheduler where * each rail will execute. *

* To merge the parallel 'rails' back into a single sequence, use {@link ParallelFlowable#sequential()}. *

* *

*
Backpressure:
*
The operator requires the upstream to honor backpressure and each 'rail' honors backpressure * as well.
*
Scheduler:
*
{@code parallel} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.0.5 - experimental; 2.1 - beta * @param parallelism the number of 'rails' to use * @param prefetch the number of items each 'rail' should prefetch * @return the new ParallelFlowable instance * @since 2.2 */ @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) @CheckReturnValue public final ParallelFlowable parallel(int parallelism, int prefetch) { ObjectHelper.verifyPositive(parallelism, "parallelism"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return ParallelFlowable.from(this, parallelism, prefetch); } /** * Returns a {@link ConnectableFlowable}, which is a variety of Publisher that waits until its * {@link ConnectableFlowable#connect connect} method is called before it begins emitting items to those * {@link Subscriber}s that have subscribed to it. *

* *

*
Backpressure:
*
The returned {@code ConnectableFlowable} honors backpressure for each of its {@code Subscriber}s * and expects the source {@code Publisher} to honor backpressure as well. If this expectation is violated, * the operator will signal a {@code MissingBackpressureException} to its {@code Subscriber}s and disconnect.
*
Scheduler:
*
{@code publish} does not operate by default on a particular {@link Scheduler}.
*
* * @return a {@link ConnectableFlowable} that upon connection causes the source Publisher to emit items * to its {@link Subscriber}s * @see ReactiveX operators documentation: Publish */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final ConnectableFlowable publish() { return publish(bufferSize()); } /** * Returns a Flowable that emits the results of invoking a specified selector on items emitted by a * {@link ConnectableFlowable} that shares a single subscription to the underlying sequence. *

* *

*
Backpressure:
*
The operator expects the source {@code Publisher} to honor backpressure and if this expectation is * violated, the operator will signal a {@code MissingBackpressureException} through the {@code Publisher} * provided to the function. Since the {@code Publisher} returned by the {@code selector} may be * independent of the provided {@code Publisher} to the function, the output's backpressure behavior * is determined by this returned {@code Publisher}.
*
Scheduler:
*
{@code publish} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the resulting Publisher * @param selector * a function that can use the multicasted source sequence as many times as needed, without * causing multiple subscriptions to the source sequence. Subscribers to the given source will * receive all notifications of the source from the time of the subscription forward. * @return a Flowable that emits the results of invoking the selector on the items emitted by a {@link ConnectableFlowable} that shares a single subscription to the underlying sequence * @see ReactiveX operators documentation: Publish */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable publish(Function, ? extends Publisher> selector) { return publish(selector, bufferSize()); } /** * Returns a Flowable that emits the results of invoking a specified selector on items emitted by a * {@link ConnectableFlowable} that shares a single subscription to the underlying sequence. *

* *

*
Backpressure:
*
The operator expects the source {@code Publisher} to honor backpressure and if this expectation is * violated, the operator will signal a {@code MissingBackpressureException} through the {@code Publisher} * provided to the function. Since the {@code Publisher} returned by the {@code selector} may be * independent of the provided {@code Publisher} to the function, the output's backpressure behavior * is determined by this returned {@code Publisher}.
*
Scheduler:
*
{@code publish} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the resulting Publisher * @param selector * a function that can use the multicasted source sequence as many times as needed, without * causing multiple subscriptions to the source sequence. Subscribers to the given source will * receive all notifications of the source from the time of the subscription forward. * @param prefetch * the number of elements to prefetch from the current Flowable * @return a Flowable that emits the results of invoking the selector on the items emitted by a {@link ConnectableFlowable} that shares a single subscription to the underlying sequence * @see ReactiveX operators documentation: Publish */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable publish(Function, ? extends Publisher> selector, int prefetch) { ObjectHelper.requireNonNull(selector, "selector is null"); ObjectHelper.verifyPositive(prefetch, "prefetch"); return RxJavaPlugins.onAssembly(new FlowablePublishMulticast(this, selector, prefetch, false)); } /** * Returns a {@link ConnectableFlowable}, which is a variety of Publisher that waits until its * {@link ConnectableFlowable#connect connect} method is called before it begins emitting items to those * {@link Subscriber}s that have subscribed to it. *

* *

*
Backpressure:
*
The returned {@code ConnectableFlowable} honors backpressure for each of its {@code Subscriber}s * and expects the source {@code Publisher} to honor backpressure as well. If this expectation is violated, * the operator will signal a {@code MissingBackpressureException} to its {@code Subscriber}s and disconnect.
*
Scheduler:
*
{@code publish} does not operate by default on a particular {@link Scheduler}.
*
* * @param bufferSize * the number of elements to prefetch from the current Flowable * @return a {@link ConnectableFlowable} that upon connection causes the source Publisher to emit items * to its {@link Subscriber}s * @see ReactiveX operators documentation: Publish */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final ConnectableFlowable publish(int bufferSize) { ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return FlowablePublish.create(this, bufferSize); } /** * Requests {@code n} initially from the upstream and then 75% of {@code n} subsequently * after 75% of {@code n} values have been emitted to the downstream. * *

This operator allows preventing the downstream to trigger unbounded mode via {@code request(Long.MAX_VALUE)} * or compensate for the per-item overhead of small and frequent requests. * *

*
Backpressure:
*
The operator expects backpressure from upstream and honors backpressure from downstream.
*
Scheduler:
*
{@code rebatchRequests} does not operate by default on a particular {@link Scheduler}.
*
* * @param n the initial request amount, further request will happen after 75% of this value * @return the Publisher that rebatches request amounts from downstream * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable rebatchRequests(int n) { return observeOn(ImmediateThinScheduler.INSTANCE, true, n); } /** * Returns a Maybe that applies a specified accumulator function to the first item emitted by a source * Publisher, then feeds the result of that function along with the second item emitted by the source * Publisher into the same function, and so on until all items have been emitted by the finite source Publisher, * and emits the final result from the final call to your function as its sole item. *

* *

* This technique, which is called "reduce" here, is sometimes called "aggregate," "fold," "accumulate," * "compress," or "inject" in other programming contexts. Groovy, for instance, has an {@code inject} method * that does a similar operation on lists. *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulator object to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure of its downstream consumer and consumes the * upstream source in unbounded mode.
*
Scheduler:
*
{@code reduce} does not operate by default on a particular {@link Scheduler}.
*
* * @param reducer * an accumulator function to be invoked on each item emitted by the source Publisher, whose * result will be used in the next accumulator call * @return a Maybe that emits a single item that is the result of accumulating the items emitted by * the source Flowable * @see ReactiveX operators documentation: Reduce * @see Wikipedia: Fold (higher-order function) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Maybe reduce(BiFunction reducer) { ObjectHelper.requireNonNull(reducer, "reducer is null"); return RxJavaPlugins.onAssembly(new FlowableReduceMaybe(this, reducer)); } /** * Returns a Single that applies a specified accumulator function to the first item emitted by a source * Publisher and a specified seed value, then feeds the result of that function along with the second item * emitted by a Publisher into the same function, and so on until all items have been emitted by the * finite source Publisher, emitting the final result from the final call to your function as its sole item. *

* *

* This technique, which is called "reduce" here, is sometimes called "aggregate," "fold," "accumulate," * "compress," or "inject" in other programming contexts. Groovy, for instance, has an {@code inject} method * that does a similar operation on lists. *

* Note that the {@code seed} is shared among all subscribers to the resulting Publisher * and may cause problems if it is mutable. To make sure each subscriber gets its own value, defer * the application of this operator via {@link #defer(Callable)}: *


     * Publisher<T> source = ...
     * Single.defer(() -> source.reduce(new ArrayList<>(), (list, item) -> list.add(item)));
     *
     * // alternatively, by using compose to stay fluent
     *
     * source.compose(o ->
     *     Flowable.defer(() -> o.reduce(new ArrayList<>(), (list, item) -> list.add(item)).toFlowable())
     * ).firstOrError();
     *
     * // or, by using reduceWith instead of reduce
     *
     * source.reduceWith(() -> new ArrayList<>(), (list, item) -> list.add(item)));
     * 
*

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulator object to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure of its downstream consumer and consumes the * upstream source in unbounded mode.
*
Scheduler:
*
{@code reduce} does not operate by default on a particular {@link Scheduler}.
*
* * @param the accumulator and output value type * @param seed * the initial (seed) accumulator value * @param reducer * an accumulator function to be invoked on each item emitted by the source Publisher, the * result of which will be used in the next accumulator call * @return a Single that emits a single item that is the result of accumulating the output from the * items emitted by the source Publisher * @see ReactiveX operators documentation: Reduce * @see Wikipedia: Fold (higher-order function) * @see #reduceWith(Callable, BiFunction) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single reduce(R seed, BiFunction reducer) { ObjectHelper.requireNonNull(seed, "seed is null"); ObjectHelper.requireNonNull(reducer, "reducer is null"); return RxJavaPlugins.onAssembly(new FlowableReduceSeedSingle(this, seed, reducer)); } /** * Returns a Single that applies a specified accumulator function to the first item emitted by a source * Publisher and a seed value derived from calling a specified seedSupplier, then feeds the result * of that function along with the second item emitted by a Publisher into the same function, and so on until * all items have been emitted by the finite source Publisher, emitting the final result from the final call to your * function as its sole item. *

* *

* This technique, which is called "reduce" here, is sometimes called "aggregate," "fold," "accumulate," * "compress," or "inject" in other programming contexts. Groovy, for instance, has an {@code inject} method * that does a similar operation on lists. *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulator object to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure of its downstream consumer and consumes the * upstream source in unbounded mode.
*
Scheduler:
*
{@code reduceWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param the accumulator and output value type * @param seedSupplier * the Callable that provides the initial (seed) accumulator value for each individual Subscriber * @param reducer * an accumulator function to be invoked on each item emitted by the source Publisher, the * result of which will be used in the next accumulator call * @return a Single that emits a single item that is the result of accumulating the output from the * items emitted by the source Publisher * @see ReactiveX operators documentation: Reduce * @see Wikipedia: Fold (higher-order function) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single reduceWith(Callable seedSupplier, BiFunction reducer) { ObjectHelper.requireNonNull(seedSupplier, "seedSupplier is null"); ObjectHelper.requireNonNull(reducer, "reducer is null"); return RxJavaPlugins.onAssembly(new FlowableReduceWithSingle(this, seedSupplier, reducer)); } /** * Returns a Flowable that repeats the sequence of items emitted by the source Publisher indefinitely. *

* *

*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code repeat} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Flowable that emits the items emitted by the source Publisher repeatedly and in sequence * @see ReactiveX operators documentation: Repeat */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable repeat() { return repeat(Long.MAX_VALUE); } /** * Returns a Flowable that repeats the sequence of items emitted by the source Publisher at most * {@code count} times. *

* *

*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code repeat} does not operate by default on a particular {@link Scheduler}.
*
* * @param times * the number of times the source Publisher items are repeated, a count of 0 will yield an empty * sequence * @return a Flowable that repeats the sequence of items emitted by the source Publisher at most * {@code count} times * @throws IllegalArgumentException * if {@code count} is less than zero * @see ReactiveX operators documentation: Repeat */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable repeat(long times) { if (times < 0) { throw new IllegalArgumentException("times >= 0 required but it was " + times); } if (times == 0) { return empty(); } return RxJavaPlugins.onAssembly(new FlowableRepeat(this, times)); } /** * Returns a Flowable that repeats the sequence of items emitted by the source Publisher until * the provided stop function returns true. *

* *

*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code repeatUntil} does not operate by default on a particular {@link Scheduler}.
*
* * @param stop * a boolean supplier that is called when the current Flowable completes and unless it returns * false, the current Flowable is resubscribed * @return the new Flowable instance * @throws NullPointerException * if {@code stop} is null * @see ReactiveX operators documentation: Repeat */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable repeatUntil(BooleanSupplier stop) { ObjectHelper.requireNonNull(stop, "stop is null"); return RxJavaPlugins.onAssembly(new FlowableRepeatUntil(this, stop)); } /** * Returns a Flowable that emits the same values as the source Publisher with the exception of an * {@code onComplete}. An {@code onComplete} notification from the source will result in the emission of * a {@code void} item to the Publisher provided as an argument to the {@code notificationHandler} * function. If that Publisher calls {@code onComplete} or {@code onError} then {@code repeatWhen} will * call {@code onComplete} or {@code onError} on the child subscription. Otherwise, this Publisher will * resubscribe to the source Publisher. *

* *

*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code repeatWhen} does not operate by default on a particular {@link Scheduler}.
*
* * @param handler * receives a Publisher of notifications with which a user can complete or error, aborting the repeat. * @return the source Publisher modified with repeat logic * @see ReactiveX operators documentation: Repeat */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable repeatWhen(final Function, ? extends Publisher> handler) { ObjectHelper.requireNonNull(handler, "handler is null"); return RxJavaPlugins.onAssembly(new FlowableRepeatWhen(this, handler)); } /** * Returns a {@link ConnectableFlowable} that shares a single subscription to the underlying Publisher * that will replay all of its items and notifications to any future {@link Subscriber}. A Connectable * Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is * subscribed to, but only when its {@code connect} method is called. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
This version of {@code replay} does not operate by default on a particular {@link Scheduler}.
*
* * @return a {@link ConnectableFlowable} that upon connection causes the source Publisher to emit its * items to its {@link Subscriber}s * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final ConnectableFlowable replay() { return FlowableReplay.createFrom(this); } /** * Returns a Flowable that emits items that are the results of invoking a specified selector on the items * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
This version of {@code replay} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the resulting Publisher * @param selector * the selector function, which can use the multicasted sequence as many times as needed, without * causing multiple subscriptions to the Publisher * @return a Flowable that emits items that are the results of invoking the selector on a * {@link ConnectableFlowable} that shares a single subscription to the source Publisher * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable replay(Function, ? extends Publisher> selector) { ObjectHelper.requireNonNull(selector, "selector is null"); return FlowableReplay.multicastSelector(FlowableInternalHelper.replayCallable(this), selector); } /** * Returns a Flowable that emits items that are the results of invoking a specified selector on items * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, * replaying {@code bufferSize} notifications. *

* Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than * {@code bufferSize} source emissions. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
This version of {@code replay} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the resulting Publisher * @param selector * the selector function, which can use the multicasted sequence as many times as needed, without * causing multiple subscriptions to the Publisher * @param bufferSize * the buffer size that limits the number of items the connectable Publisher can replay * @return a Flowable that emits items that are the results of invoking the selector on items emitted by * a {@link ConnectableFlowable} that shares a single subscription to the source Publisher * replaying no more than {@code bufferSize} items * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable replay(Function, ? extends Publisher> selector, final int bufferSize) { ObjectHelper.requireNonNull(selector, "selector is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return FlowableReplay.multicastSelector(FlowableInternalHelper.replayCallable(this, bufferSize), selector); } /** * Returns a Flowable that emits items that are the results of invoking a specified selector on items * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, * replaying no more than {@code bufferSize} items that were emitted within a specified time window. *

* Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than * {@code bufferSize} source emissions. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
This version of {@code replay} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param * the type of items emitted by the resulting Publisher * @param selector * a selector function, which can use the multicasted sequence as many times as needed, without * causing multiple subscriptions to the Publisher * @param bufferSize * the buffer size that limits the number of items the connectable Publisher can replay * @param time * the duration of the window in which the replayed items must have been emitted * @param unit * the time unit of {@code time} * @return a Flowable that emits items that are the results of invoking the selector on items emitted by * a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, and * replays no more than {@code bufferSize} items that were emitted within the window defined by * {@code time} * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable replay(Function, ? extends Publisher> selector, int bufferSize, long time, TimeUnit unit) { return replay(selector, bufferSize, time, unit, Schedulers.computation()); } /** * Returns a Flowable that emits items that are the results of invoking a specified selector on items * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, * replaying no more than {@code bufferSize} items that were emitted within a specified time window. *

* Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than * {@code bufferSize} source emissions. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param * the type of items emitted by the resulting Publisher * @param selector * a selector function, which can use the multicasted sequence as many times as needed, without * causing multiple subscriptions to the Publisher * @param bufferSize * the buffer size that limits the number of items the connectable Publisher can replay * @param time * the duration of the window in which the replayed items must have been emitted * @param unit * the time unit of {@code time} * @param scheduler * the Scheduler that is the time source for the window * @return a Flowable that emits items that are the results of invoking the selector on items emitted by * a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, and * replays no more than {@code bufferSize} items that were emitted within the window defined by * {@code time} * @throws IllegalArgumentException * if {@code bufferSize} is less than zero * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable replay(Function, ? extends Publisher> selector, final int bufferSize, final long time, final TimeUnit unit, final Scheduler scheduler) { ObjectHelper.requireNonNull(selector, "selector is null"); ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return FlowableReplay.multicastSelector( FlowableInternalHelper.replayCallable(this, bufferSize, time, unit, scheduler), selector); } /** * Returns a Flowable that emits items that are the results of invoking a specified selector on items * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, * replaying a maximum of {@code bufferSize} items. *

* Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than * {@code bufferSize} source emissions. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param * the type of items emitted by the resulting Publisher * @param selector * a selector function, which can use the multicasted sequence as many times as needed, without * causing multiple subscriptions to the Publisher * @param bufferSize * the buffer size that limits the number of items the connectable Publisher can replay * @param scheduler * the Scheduler on which the replay is observed * @return a Flowable that emits items that are the results of invoking the selector on items emitted by * a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, * replaying no more than {@code bufferSize} notifications * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable replay(final Function, ? extends Publisher> selector, final int bufferSize, final Scheduler scheduler) { ObjectHelper.requireNonNull(selector, "selector is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return FlowableReplay.multicastSelector(FlowableInternalHelper.replayCallable(this, bufferSize), FlowableInternalHelper.replayFunction(selector, scheduler) ); } /** * Returns a Flowable that emits items that are the results of invoking a specified selector on items * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, * replaying all items that were emitted within a specified time window. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
This version of {@code replay} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param * the type of items emitted by the resulting Publisher * @param selector * a selector function, which can use the multicasted sequence as many times as needed, without * causing multiple subscriptions to the Publisher * @param time * the duration of the window in which the replayed items must have been emitted * @param unit * the time unit of {@code time} * @return a Flowable that emits items that are the results of invoking the selector on items emitted by * a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, * replaying all items that were emitted within the window defined by {@code time} * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable replay(Function, ? extends Publisher> selector, long time, TimeUnit unit) { return replay(selector, time, unit, Schedulers.computation()); } /** * Returns a Flowable that emits items that are the results of invoking a specified selector on items * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, * replaying all items that were emitted within a specified time window. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param * the type of items emitted by the resulting Publisher * @param selector * a selector function, which can use the multicasted sequence as many times as needed, without * causing multiple subscriptions to the Publisher * @param time * the duration of the window in which the replayed items must have been emitted * @param unit * the time unit of {@code time} * @param scheduler * the scheduler that is the time source for the window * @return a Flowable that emits items that are the results of invoking the selector on items emitted by * a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, * replaying all items that were emitted within the window defined by {@code time} * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable replay(Function, ? extends Publisher> selector, final long time, final TimeUnit unit, final Scheduler scheduler) { ObjectHelper.requireNonNull(selector, "selector is null"); ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return FlowableReplay.multicastSelector(FlowableInternalHelper.replayCallable(this, time, unit, scheduler), selector); } /** * Returns a Flowable that emits items that are the results of invoking a specified selector on items * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param * the type of items emitted by the resulting Publisher * @param selector * a selector function, which can use the multicasted sequence as many times as needed, without * causing multiple subscriptions to the Publisher * @param scheduler * the Scheduler where the replay is observed * @return a Flowable that emits items that are the results of invoking the selector on items emitted by * a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, * replaying all items * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable replay(final Function, ? extends Publisher> selector, final Scheduler scheduler) { ObjectHelper.requireNonNull(selector, "selector is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return FlowableReplay.multicastSelector(FlowableInternalHelper.replayCallable(this), FlowableInternalHelper.replayFunction(selector, scheduler)); } /** * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher that * replays at most {@code bufferSize} items emitted by that Publisher. A Connectable Publisher resembles * an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only * when its {@code connect} method is called. *

* Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than * {@code bufferSize} source emissions. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
This version of {@code replay} does not operate by default on a particular {@link Scheduler}.
*
* * @param bufferSize * the buffer size that limits the number of items that can be replayed * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * replays at most {@code bufferSize} items emitted by that Publisher * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final ConnectableFlowable replay(final int bufferSize) { ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return FlowableReplay.create(this, bufferSize); } /** * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * replays at most {@code bufferSize} items that were emitted during a specified time window. A Connectable * Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is * subscribed to, but only when its {@code connect} method is called. *

* Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than * {@code bufferSize} source emissions. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
This version of {@code replay} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param bufferSize * the buffer size that limits the number of items that can be replayed * @param time * the duration of the window in which the replayed items must have been emitted * @param unit * the time unit of {@code time} * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * replays at most {@code bufferSize} items that were emitted during the window defined by * {@code time} * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final ConnectableFlowable replay(int bufferSize, long time, TimeUnit unit) { return replay(bufferSize, time, unit, Schedulers.computation()); } /** * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * that replays a maximum of {@code bufferSize} items that are emitted within a specified time window. A * Connectable Publisher resembles an ordinary Publisher, except that it does not begin emitting items * when it is subscribed to, but only when its {@code connect} method is called. *

* Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than * {@code bufferSize} source emissions. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param bufferSize * the buffer size that limits the number of items that can be replayed * @param time * the duration of the window in which the replayed items must have been emitted * @param unit * the time unit of {@code time} * @param scheduler * the scheduler that is used as a time source for the window * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * replays at most {@code bufferSize} items that were emitted during the window defined by * {@code time} * @throws IllegalArgumentException * if {@code bufferSize} is less than zero * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final ConnectableFlowable replay(final int bufferSize, final long time, final TimeUnit unit, final Scheduler scheduler) { ObjectHelper.verifyPositive(bufferSize, "bufferSize"); ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return FlowableReplay.create(this, time, unit, scheduler, bufferSize); } /** * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * replays at most {@code bufferSize} items emitted by that Publisher. A Connectable Publisher resembles * an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only * when its {@code connect} method is called. *

* Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than * {@code bufferSize} source emissions. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param bufferSize * the buffer size that limits the number of items that can be replayed * @param scheduler * the scheduler on which the Subscribers will observe the emitted items * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * replays at most {@code bufferSize} items that were emitted by the Publisher * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final ConnectableFlowable replay(final int bufferSize, final Scheduler scheduler) { ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return FlowableReplay.observeOn(replay(bufferSize), scheduler); } /** * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * replays all items emitted by that Publisher within a specified time window. A Connectable Publisher * resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, * but only when its {@code connect} method is called. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
This version of {@code replay} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param time * the duration of the window in which the replayed items must have been emitted * @param unit * the time unit of {@code time} * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * replays the items that were emitted during the window defined by {@code time} * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final ConnectableFlowable replay(long time, TimeUnit unit) { return replay(time, unit, Schedulers.computation()); } /** * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * replays all items emitted by that Publisher within a specified time window. A Connectable Publisher * resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, * but only when its {@code connect} method is called. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param time * the duration of the window in which the replayed items must have been emitted * @param unit * the time unit of {@code time} * @param scheduler * the Scheduler that is the time source for the window * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and * replays the items that were emitted during the window defined by {@code time} * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final ConnectableFlowable replay(final long time, final TimeUnit unit, final Scheduler scheduler) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return FlowableReplay.create(this, time, unit, scheduler); } /** * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher that * will replay all of its items and notifications to any future {@link Subscriber} on the given * {@link Scheduler}. A Connectable Publisher resembles an ordinary Publisher, except that it does not * begin emitting items when it is subscribed to, but only when its {@code connect} method is called. *

* *

*
Backpressure:
*
This operator supports backpressure. Note that the upstream requests are determined by the child * Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will * request 100 elements from the underlying Publisher sequence.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param scheduler * the Scheduler on which the Subscribers will observe the emitted items * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher that * will replay all of its items and notifications to any future {@link Subscriber} on the given * {@link Scheduler} * @see ReactiveX operators documentation: Replay */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final ConnectableFlowable replay(final Scheduler scheduler) { ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return FlowableReplay.observeOn(replay(), scheduler); } /** * Returns a Flowable that mirrors the source Publisher, resubscribing to it if it calls {@code onError} * (infinite retry count). *

* *

* If the source Publisher calls {@link Subscriber#onError}, this method will resubscribe to the source * Publisher rather than propagating the {@code onError} call. *

* Any and all items emitted by the source Publisher will be emitted by the resulting Publisher, even * those emitted during failed subscriptions. For example, if a Publisher fails at first but emits * {@code [1, 2]} then succeeds the second time and emits {@code [1, 2, 3, 4, 5]} then the complete sequence * of emissions and notifications would be {@code [1, 2, 1, 2, 3, 4, 5, onComplete]}. *

*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code retry} does not operate by default on a particular {@link Scheduler}.
*
* * @return the source Publisher modified with retry logic * @see ReactiveX operators documentation: Retry */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable retry() { return retry(Long.MAX_VALUE, Functions.alwaysTrue()); } /** * Returns a Flowable that mirrors the source Publisher, resubscribing to it if it calls {@code onError} * and the predicate returns true for that specific exception and retry count. *

* *

*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code retry} does not operate by default on a particular {@link Scheduler}.
*
* * @param predicate * the predicate that determines if a resubscription may happen in case of a specific exception * and retry count * @return the source Publisher modified with retry logic * @see #retry() * @see ReactiveX operators documentation: Retry */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable retry(BiPredicate predicate) { ObjectHelper.requireNonNull(predicate, "predicate is null"); return RxJavaPlugins.onAssembly(new FlowableRetryBiPredicate(this, predicate)); } /** * Returns a Flowable that mirrors the source Publisher, resubscribing to it if it calls {@code onError} * up to a specified number of retries. *

* *

* If the source Publisher calls {@link Subscriber#onError}, this method will resubscribe to the source * Publisher for a maximum of {@code count} resubscriptions rather than propagating the * {@code onError} call. *

* Any and all items emitted by the source Publisher will be emitted by the resulting Publisher, even * those emitted during failed subscriptions. For example, if a Publisher fails at first but emits * {@code [1, 2]} then succeeds the second time and emits {@code [1, 2, 3, 4, 5]} then the complete sequence * of emissions and notifications would be {@code [1, 2, 1, 2, 3, 4, 5, onComplete]}. *

*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code retry} does not operate by default on a particular {@link Scheduler}.
*
* * @param count * the number of times to resubscribe if the current Flowable fails * @return the source Publisher modified with retry logic * @see ReactiveX operators documentation: Retry */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable retry(long count) { return retry(count, Functions.alwaysTrue()); } /** * Retries at most times or until the predicate returns false, whichever happens first. * *
*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code retry} does not operate by default on a particular {@link Scheduler}.
*
* @param times the number of times to resubscribe if the current Flowable fails * @param predicate the predicate called with the failure Throwable and should return true to trigger a retry. * @return the new Flowable instance */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable retry(long times, Predicate predicate) { if (times < 0) { throw new IllegalArgumentException("times >= 0 required but it was " + times); } ObjectHelper.requireNonNull(predicate, "predicate is null"); return RxJavaPlugins.onAssembly(new FlowableRetryPredicate(this, times, predicate)); } /** * Retries the current Flowable if the predicate returns true. *
*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code retry} does not operate by default on a particular {@link Scheduler}.
*
* * @param predicate the predicate that receives the failure Throwable and should return true to trigger a retry. * @return the new Flowable instance */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable retry(Predicate predicate) { return retry(Long.MAX_VALUE, predicate); } /** * Retries until the given stop function returns true. *
*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code retryUntil} does not operate by default on a particular {@link Scheduler}.
*
* @param stop the function that should return true to stop retrying * @return the new Flowable instance */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable retryUntil(final BooleanSupplier stop) { ObjectHelper.requireNonNull(stop, "stop is null"); return retry(Long.MAX_VALUE, Functions.predicateReverseFor(stop)); } /** * Returns a Flowable that emits the same values as the source Publisher with the exception of an * {@code onError}. An {@code onError} notification from the source will result in the emission of a * {@link Throwable} item to the Publisher provided as an argument to the {@code notificationHandler} * function. If that Publisher calls {@code onComplete} or {@code onError} then {@code retry} will call * {@code onComplete} or {@code onError} on the child subscription. Otherwise, this Publisher will * resubscribe to the source Publisher. *

* *

* Example: * * This retries 3 times, each time incrementing the number of seconds it waits. * *


     *  Flowable.create((FlowableEmitter<? super String> s) -> {
     *      System.out.println("subscribing");
     *      s.onError(new RuntimeException("always fails"));
     *  }, BackpressureStrategy.BUFFER).retryWhen(attempts -> {
     *      return attempts.zipWith(Flowable.range(1, 3), (n, i) -> i).flatMap(i -> {
     *          System.out.println("delay retry by " + i + " second(s)");
     *          return Flowable.timer(i, TimeUnit.SECONDS);
     *      });
     *  }).blockingForEach(System.out::println);
     * 
* * Output is: * *
 {@code
     * subscribing
     * delay retry by 1 second(s)
     * subscribing
     * delay retry by 2 second(s)
     * subscribing
     * delay retry by 3 second(s)
     * subscribing
     * } 
*

* Note that the inner {@code Publisher} returned by the handler function should signal * either {@code onNext}, {@code onError} or {@code onComplete} in response to the received * {@code Throwable} to indicate the operator should retry or terminate. If the upstream to * the operator is asynchronous, signaling onNext followed by onComplete immediately may * result in the sequence to be completed immediately. Similarly, if this inner * {@code Publisher} signals {@code onError} or {@code onComplete} while the upstream is * active, the sequence is terminated with the same signal immediately. *

* The following example demonstrates how to retry an asynchronous source with a delay: *


     * Flowable.timer(1, TimeUnit.SECONDS)
     *     .doOnSubscribe(s -> System.out.println("subscribing"))
     *     .map(v -> { throw new RuntimeException(); })
     *     .retryWhen(errors -> {
     *         AtomicInteger counter = new AtomicInteger();
     *         return errors
     *                   .takeWhile(e -> counter.getAndIncrement() != 3)
     *                   .flatMap(e -> {
     *                       System.out.println("delay retry by " + counter.get() + " second(s)");
     *                       return Flowable.timer(counter.get(), TimeUnit.SECONDS);
     *                   });
     *     })
     *     .blockingSubscribe(System.out::println, System.out::println);
     * 
*
*
Backpressure:
*
The operator honors downstream backpressure and expects both the source * and inner {@code Publisher}s to honor backpressure as well. * If this expectation is violated, the operator may throw an {@code IllegalStateException}.
*
Scheduler:
*
{@code retryWhen} does not operate by default on a particular {@link Scheduler}.
*
* * @param handler * receives a Publisher of notifications with which a user can complete or error, aborting the * retry * @return the source Publisher modified with retry logic * @see ReactiveX operators documentation: Retry */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable retryWhen( final Function, ? extends Publisher> handler) { ObjectHelper.requireNonNull(handler, "handler is null"); return RxJavaPlugins.onAssembly(new FlowableRetryWhen(this, handler)); } /** * Subscribes to the current Flowable and wraps the given Subscriber into a SafeSubscriber * (if not already a SafeSubscriber) that * deals with exceptions thrown by a misbehaving Subscriber (that doesn't follow the * Reactive Streams specification). *
*
Backpressure:
*
This operator leaves the reactive world and the backpressure behavior depends on the Subscriber's behavior.
*
Scheduler:
*
{@code safeSubscribe} does not operate by default on a particular {@link Scheduler}.
*
* @param s the incoming Subscriber instance * @throws NullPointerException if s is null */ @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final void safeSubscribe(Subscriber s) { ObjectHelper.requireNonNull(s, "s is null"); if (s instanceof SafeSubscriber) { subscribe((SafeSubscriber)s); } else { subscribe(new SafeSubscriber(s)); } } /** * Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher * within periodic time intervals. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
{@code sample} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param period * the sampling rate * @param unit * the {@link TimeUnit} in which {@code period} is defined * @return a Flowable that emits the results of sampling the items emitted by the source Publisher at * the specified time interval * @see ReactiveX operators documentation: Sample * @see RxJava wiki: Backpressure * @see #throttleLast(long, TimeUnit) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable sample(long period, TimeUnit unit) { return sample(period, unit, Schedulers.computation()); } /** * Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher * within periodic time intervals and optionally emit the very last upstream item when the upstream completes. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
{@code sample} operates by default on the {@code computation} {@link Scheduler}.
*
* *

History: 2.0.5 - experimental * @param period * the sampling rate * @param unit * the {@link TimeUnit} in which {@code period} is defined * @param emitLast * if true and the upstream completes while there is still an unsampled item available, * that item is emitted to downstream before completion * if false, an unsampled last item is ignored. * @return a Flowable that emits the results of sampling the items emitted by the source Publisher at * the specified time interval * @see ReactiveX operators documentation: Sample * @see RxJava wiki: Backpressure * @see #throttleLast(long, TimeUnit) * @since 2.1 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable sample(long period, TimeUnit unit, boolean emitLast) { return sample(period, unit, Schedulers.computation(), emitLast); } /** * Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher * within periodic time intervals, where the intervals are defined on a particular Scheduler. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param period * the sampling rate * @param unit * the {@link TimeUnit} in which {@code period} is defined * @param scheduler * the {@link Scheduler} to use when sampling * @return a Flowable that emits the results of sampling the items emitted by the source Publisher at * the specified time interval * @see ReactiveX operators documentation: Sample * @see RxJava wiki: Backpressure * @see #throttleLast(long, TimeUnit, Scheduler) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable sample(long period, TimeUnit unit, Scheduler scheduler) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableSampleTimed(this, period, unit, scheduler, false)); } /** * Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher * within periodic time intervals, where the intervals are defined on a particular Scheduler * and optionally emit the very last upstream item when the upstream completes. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* *

History: 2.0.5 - experimental * @param period * the sampling rate * @param unit * the {@link TimeUnit} in which {@code period} is defined * @param scheduler * the {@link Scheduler} to use when sampling * @param emitLast * if true and the upstream completes while there is still an unsampled item available, * that item is emitted to downstream before completion * if false, an unsampled last item is ignored. * @return a Flowable that emits the results of sampling the items emitted by the source Publisher at * the specified time interval * @see ReactiveX operators documentation: Sample * @see RxJava wiki: Backpressure * @see #throttleLast(long, TimeUnit, Scheduler) * @since 2.1 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable sample(long period, TimeUnit unit, Scheduler scheduler, boolean emitLast) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableSampleTimed(this, period, unit, scheduler, emitLast)); } /** * Returns a Flowable that, when the specified {@code sampler} Publisher emits an item or completes, * emits the most recently emitted item (if any) emitted by the source Publisher since the previous * emission from the {@code sampler} Publisher. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses the emissions of the {@code sampler} * Publisher to control data flow.
*
Scheduler:
*
This version of {@code sample} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the sampler Publisher * @param sampler * the Publisher to use for sampling the source Publisher * @return a Flowable that emits the results of sampling the items emitted by this Publisher whenever * the {@code sampler} Publisher emits an item or completes * @see ReactiveX operators documentation: Sample * @see RxJava wiki: Backpressure */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable sample(Publisher sampler) { ObjectHelper.requireNonNull(sampler, "sampler is null"); return RxJavaPlugins.onAssembly(new FlowableSamplePublisher(this, sampler, false)); } /** * Returns a Flowable that, when the specified {@code sampler} Publisher emits an item or completes, * emits the most recently emitted item (if any) emitted by the source Publisher since the previous * emission from the {@code sampler} Publisher * and optionally emit the very last upstream item when the upstream or other Publisher complete. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses the emissions of the {@code sampler} * Publisher to control data flow.
*
Scheduler:
*
This version of {@code sample} does not operate by default on a particular {@link Scheduler}.
*
* *

History: 2.0.5 - experimental * @param the element type of the sampler Publisher * @param sampler * the Publisher to use for sampling the source Publisher * @param emitLast * if true and the upstream completes while there is still an unsampled item available, * that item is emitted to downstream before completion * if false, an unsampled last item is ignored. * @return a Flowable that emits the results of sampling the items emitted by this Publisher whenever * the {@code sampler} Publisher emits an item or completes * @see ReactiveX operators documentation: Sample * @see RxJava wiki: Backpressure * @since 2.1 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable sample(Publisher sampler, boolean emitLast) { ObjectHelper.requireNonNull(sampler, "sampler is null"); return RxJavaPlugins.onAssembly(new FlowableSamplePublisher(this, sampler, emitLast)); } /** * Returns a Flowable that applies a specified accumulator function to the first item emitted by a source * Publisher, then feeds the result of that function along with the second item emitted by the source * Publisher into the same function, and so on until all items have been emitted by the source Publisher, * emitting the result of each of these iterations. *

* *

* This sort of function is sometimes called an accumulator. *

*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * Violating this expectation, a {@code MissingBackpressureException} may get signaled somewhere downstream.
*
Scheduler:
*
{@code scan} does not operate by default on a particular {@link Scheduler}.
*
* * @param accumulator * an accumulator function to be invoked on each item emitted by the source Publisher, whose * result will be emitted to {@link Subscriber}s via {@link Subscriber#onNext onNext} and used in the * next accumulator call * @return a Flowable that emits the results of each call to the accumulator function * @see ReactiveX operators documentation: Scan */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable scan(BiFunction accumulator) { ObjectHelper.requireNonNull(accumulator, "accumulator is null"); return RxJavaPlugins.onAssembly(new FlowableScan(this, accumulator)); } /** * Returns a Flowable that applies a specified accumulator function to the first item emitted by a source * Publisher and a seed value, then feeds the result of that function along with the second item emitted by * the source Publisher into the same function, and so on until all items have been emitted by the source * Publisher, emitting the result of each of these iterations. *

* *

* This sort of function is sometimes called an accumulator. *

* Note that the Publisher that results from this method will emit {@code initialValue} as its first * emitted item. *

* Note that the {@code initialValue} is shared among all subscribers to the resulting Publisher * and may cause problems if it is mutable. To make sure each subscriber gets its own value, defer * the application of this operator via {@link #defer(Callable)}: *


     * Publisher<T> source = ...
     * Flowable.defer(() -> source.scan(new ArrayList<>(), (list, item) -> list.add(item)));
     *
     * // alternatively, by using compose to stay fluent
     *
     * source.compose(o ->
     *     Flowable.defer(() -> o.scan(new ArrayList<>(), (list, item) -> list.add(item)))
     * );
     * 
*
*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * Violating this expectation, a {@code MissingBackpressureException} may get signaled somewhere downstream.
*
Scheduler:
*
{@code scan} does not operate by default on a particular {@link Scheduler}.
*
* * @param the initial, accumulator and result type * @param initialValue * the initial (seed) accumulator item * @param accumulator * an accumulator function to be invoked on each item emitted by the source Publisher, whose * result will be emitted to {@link Subscriber}s via {@link Subscriber#onNext onNext} and used in the * next accumulator call * @return a Flowable that emits {@code initialValue} followed by the results of each call to the * accumulator function * @see ReactiveX operators documentation: Scan */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable scan(final R initialValue, BiFunction accumulator) { ObjectHelper.requireNonNull(initialValue, "initialValue is null"); return scanWith(Functions.justCallable(initialValue), accumulator); } /** * Returns a Flowable that applies a specified accumulator function to the first item emitted by a source * Publisher and a seed value, then feeds the result of that function along with the second item emitted by * the source Publisher into the same function, and so on until all items have been emitted by the source * Publisher, emitting the result of each of these iterations. *

* *

* This sort of function is sometimes called an accumulator. *

* Note that the Publisher that results from this method will emit the value returned by * the {@code seedSupplier} as its first item. *

*
Backpressure:
*
The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well. * Violating this expectation, a {@code MissingBackpressureException} may get signaled somewhere downstream.
*
Scheduler:
*
{@code scanWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param the initial, accumulator and result type * @param seedSupplier * a Callable that returns the initial (seed) accumulator item for each individual Subscriber * @param accumulator * an accumulator function to be invoked on each item emitted by the source Publisher, whose * result will be emitted to {@link Subscriber}s via {@link Subscriber#onNext onNext} and used in the * next accumulator call * @return a Flowable that emits {@code initialValue} followed by the results of each call to the * accumulator function * @see ReactiveX operators documentation: Scan */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable scanWith(Callable seedSupplier, BiFunction accumulator) { ObjectHelper.requireNonNull(seedSupplier, "seedSupplier is null"); ObjectHelper.requireNonNull(accumulator, "accumulator is null"); return RxJavaPlugins.onAssembly(new FlowableScanSeed(this, seedSupplier, accumulator)); } /** * Forces a Publisher's emissions and notifications to be serialized and for it to obey * the Publisher contract in other ways. *

* It is possible for a Publisher to invoke its Subscribers' methods asynchronously, perhaps from * different threads. This could make such a Publisher poorly-behaved, in that it might try to invoke * {@code onComplete} or {@code onError} before one of its {@code onNext} invocations, or it might call * {@code onNext} from two different threads concurrently. You can force such a Publisher to be * well-behaved and sequential by applying the {@code serialize} method to it. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code serialize} does not operate by default on a particular {@link Scheduler}.
*
* * @return a {@link Publisher} that is guaranteed to be well-behaved and to make only serialized calls to * its Subscribers * @see ReactiveX operators documentation: Serialize */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable serialize() { return RxJavaPlugins.onAssembly(new FlowableSerialized(this)); } /** * Returns a new {@link Publisher} that multicasts (and shares a single subscription to) the original {@link Publisher}. As long as * there is at least one {@link Subscriber} this {@link Publisher} will be subscribed and emitting data. * When all subscribers have canceled it will cancel the source {@link Publisher}. *

* This is an alias for {@link #publish()}.{@link ConnectableFlowable#refCount() refCount()}. *

* *

*
Backpressure:
*
The operator honors backpressure and expects the source {@code Publisher} to honor backpressure as well. * If this expectation is violated, the operator will signal a {@code MissingBackpressureException} to * its {@code Subscriber}s.
*
Scheduler:
*
{@code share} does not operate by default on a particular {@link Scheduler}.
*
* * @return a {@code Publisher} that upon connection causes the source {@code Publisher} to emit items * to its {@link Subscriber}s * @see ReactiveX operators documentation: RefCount */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable share() { return publish().refCount(); } /** * Returns a Maybe that completes if this Flowable is empty, signals one item if this Flowable * signals exactly one item or signals an {@code IllegalArgumentException} if this Flowable signals * more than one item. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code singleElement} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Maybe that emits the single item emitted by the source Publisher * @see ReactiveX operators documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Maybe singleElement() { return RxJavaPlugins.onAssembly(new FlowableSingleMaybe(this)); } /** * Returns a Single that emits the single item emitted by the source Publisher, if that Publisher * emits only a single item, or a default item if the source Publisher emits no items. If the source * Publisher emits more than one item, an {@code IllegalArgumentException} is signaled instead. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code single} does not operate by default on a particular {@link Scheduler}.
*
* * @param defaultItem * a default value to emit if the source Publisher emits no item * @return a Single that emits the single item emitted by the source Publisher, or a default item if * the source Publisher is empty * @see ReactiveX operators documentation: First */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single single(T defaultItem) { ObjectHelper.requireNonNull(defaultItem, "defaultItem is null"); return RxJavaPlugins.onAssembly(new FlowableSingleSingle(this, defaultItem)); } /** * Returns a Single that emits the single item emitted by this Flowable, if this Flowable * emits only a single item, otherwise * if this Flowable completes without emitting any items a {@link NoSuchElementException} will be signaled and * if this Flowable emits more than one item, an {@code IllegalArgumentException} will be signaled. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure).
*
Scheduler:
*
{@code singleOrError} does not operate by default on a particular {@link Scheduler}.
*
* * @return the new Single instance * @see ReactiveX operators documentation: First */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single singleOrError() { return RxJavaPlugins.onAssembly(new FlowableSingleSingle(this, null)); } /** * Returns a Flowable that skips the first {@code count} items emitted by the source Publisher and emits * the remainder. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
This version of {@code skip} does not operate by default on a particular {@link Scheduler}.
*
* * @param count * the number of items to skip * @return a Flowable that is identical to the source Publisher except that it does not emit the first * {@code count} items that the source Publisher emits * @see ReactiveX operators documentation: Skip */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable skip(long count) { if (count <= 0L) { return RxJavaPlugins.onAssembly(this); } return RxJavaPlugins.onAssembly(new FlowableSkip(this, count)); } /** * Returns a Flowable that skips values emitted by the source Publisher before a specified time window * elapses. *

* *

*
Backpressure:
*
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and * thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code skip} does not operate on any particular scheduler but uses the current time * from the {@code computation} {@link Scheduler}.
*
* * @param time * the length of the time window to skip * @param unit * the time unit of {@code time} * @return a Flowable that skips values emitted by the source Publisher before the time window defined * by {@code time} elapses and the emits the remainder * @see ReactiveX operators documentation: Skip */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable skip(long time, TimeUnit unit) { return skipUntil(timer(time, unit)); } /** * Returns a Flowable that skips values emitted by the source Publisher before a specified time window * on a specified {@link Scheduler} elapses. *

* *

*
Backpressure:
*
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and * thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use for the timed skipping
*
* * @param time * the length of the time window to skip * @param unit * the time unit of {@code time} * @param scheduler * the {@link Scheduler} on which the timed wait happens * @return a Flowable that skips values emitted by the source Publisher before the time window defined * by {@code time} and {@code scheduler} elapses, and then emits the remainder * @see ReactiveX operators documentation: Skip */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable skip(long time, TimeUnit unit, Scheduler scheduler) { return skipUntil(timer(time, unit, scheduler)); } /** * Returns a Flowable that drops a specified number of items from the end of the sequence emitted by the * source Publisher. *

* *

* This Subscriber accumulates a queue long enough to store the first {@code count} items. As more items are * received, items are taken from the front of the queue and emitted by the returned Publisher. This causes * such items to be delayed. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
This version of {@code skipLast} does not operate by default on a particular {@link Scheduler}.
*
* * @param count * number of items to drop from the end of the source sequence * @return a Flowable that emits the items emitted by the source Publisher except for the dropped ones * at the end * @throws IndexOutOfBoundsException * if {@code count} is less than zero * @see ReactiveX operators documentation: SkipLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable skipLast(int count) { if (count < 0) { throw new IndexOutOfBoundsException("count >= 0 required but it was " + count); } if (count == 0) { return RxJavaPlugins.onAssembly(this); } return RxJavaPlugins.onAssembly(new FlowableSkipLast(this, count)); } /** * Returns a Flowable that drops items emitted by the source Publisher during a specified time window * before the source completes. *

* *

* Note: this action will cache the latest items arriving in the specified time window. *

*
Backpressure:
*
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and * thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code skipLast} does not operate on any particular scheduler but uses the current time * from the {@code computation} {@link Scheduler}.
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @return a Flowable that drops those items emitted by the source Publisher in a time window before the * source completes defined by {@code time} * @see ReactiveX operators documentation: SkipLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable skipLast(long time, TimeUnit unit) { return skipLast(time, unit, Schedulers.computation(), false, bufferSize()); } /** * Returns a Flowable that drops items emitted by the source Publisher during a specified time window * before the source completes. *

* *

* Note: this action will cache the latest items arriving in the specified time window. *

*
Backpressure:
*
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and * thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).
*
Scheduler:
*
{@code skipLast} does not operate on any particular scheduler but uses the current time * from the {@code computation} {@link Scheduler}.
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param delayError * if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed * by the downstream; if false, an exception is immediately signaled and all regular elements dropped * @return a Flowable that drops those items emitted by the source Publisher in a time window before the * source completes defined by {@code time} * @see ReactiveX operators documentation: SkipLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable skipLast(long time, TimeUnit unit, boolean delayError) { return skipLast(time, unit, Schedulers.computation(), delayError, bufferSize()); } /** * Returns a Flowable that drops items emitted by the source Publisher during a specified time window * (defined on a specified scheduler) before the source completes. *

* *

* Note: this action will cache the latest items arriving in the specified time window. *

*
Backpressure:
*
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and * thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use for tracking the current time
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param scheduler * the scheduler used as the time source * @return a Flowable that drops those items emitted by the source Publisher in a time window before the * source completes defined by {@code time} and {@code scheduler} * @see ReactiveX operators documentation: SkipLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable skipLast(long time, TimeUnit unit, Scheduler scheduler) { return skipLast(time, unit, scheduler, false, bufferSize()); } /** * Returns a Flowable that drops items emitted by the source Publisher during a specified time window * (defined on a specified scheduler) before the source completes. *

* *

* Note: this action will cache the latest items arriving in the specified time window. *

*
Backpressure:
*
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and * thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use to track the current time
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param scheduler * the scheduler used as the time source * @param delayError * if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed * by the downstream; if false, an exception is immediately signaled and all regular elements dropped * @return a Flowable that drops those items emitted by the source Publisher in a time window before the * source completes defined by {@code time} and {@code scheduler} * @see ReactiveX operators documentation: SkipLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable skipLast(long time, TimeUnit unit, Scheduler scheduler, boolean delayError) { return skipLast(time, unit, scheduler, delayError, bufferSize()); } /** * Returns a Flowable that drops items emitted by the source Publisher during a specified time window * (defined on a specified scheduler) before the source completes. *

* *

* Note: this action will cache the latest items arriving in the specified time window. *

*
Backpressure:
*
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and * thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param scheduler * the scheduler used as the time source * @param delayError * if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed * by the downstream; if false, an exception is immediately signaled and all regular elements dropped * @param bufferSize * the hint about how many elements to expect to be skipped * @return a Flowable that drops those items emitted by the source Publisher in a time window before the * source completes defined by {@code time} and {@code scheduler} * @see ReactiveX operators documentation: SkipLast */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable skipLast(long time, TimeUnit unit, Scheduler scheduler, boolean delayError, int bufferSize) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); // the internal buffer holds pairs of (timestamp, value) so double the default buffer size int s = bufferSize << 1; return RxJavaPlugins.onAssembly(new FlowableSkipLastTimed(this, time, unit, scheduler, s, delayError)); } /** * Returns a Flowable that skips items emitted by the source Publisher until a second Publisher emits * an item. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code skipUntil} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the other Publisher * @param other * the second Publisher that has to emit an item before the source Publisher's elements begin * to be mirrored by the resulting Publisher * @return a Flowable that skips items from the source Publisher until the second Publisher emits an * item, then emits the remaining items * @see ReactiveX operators documentation: SkipUntil */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable skipUntil(Publisher other) { ObjectHelper.requireNonNull(other, "other is null"); return RxJavaPlugins.onAssembly(new FlowableSkipUntil(this, other)); } /** * Returns a Flowable that skips all items emitted by the source Publisher as long as a specified * condition holds true, but emits all further source items as soon as the condition becomes false. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code skipWhile} does not operate by default on a particular {@link Scheduler}.
*
* * @param predicate * a function to test each item emitted from the source Publisher * @return a Flowable that begins emitting items emitted by the source Publisher when the specified * predicate becomes false * @see ReactiveX operators documentation: SkipWhile */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable skipWhile(Predicate predicate) { ObjectHelper.requireNonNull(predicate, "predicate is null"); return RxJavaPlugins.onAssembly(new FlowableSkipWhile(this, predicate)); } /** * Returns a Flowable that emits the events emitted by source Publisher, in a * sorted order. Each item emitted by the Publisher must implement {@link Comparable} with respect to all * other items in the sequence. * *

If any item emitted by this Flowable does not implement {@link Comparable} with respect to * all other items emitted by this Flowable, no items will be emitted and the * sequence is terminated with a {@link ClassCastException}. *

Note that calling {@code sorted} with long, non-terminating or infinite sources * might cause {@link OutOfMemoryError} * *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code sorted} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Flowable that emits the items emitted by the source Publisher in sorted order */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable sorted() { return toList().toFlowable().map(Functions.listSorter(Functions.naturalComparator())).flatMapIterable(Functions.>identity()); } /** * Returns a Flowable that emits the events emitted by source Publisher, in a * sorted order based on a specified comparison function. * *

Note that calling {@code sorted} with long, non-terminating or infinite sources * might cause {@link OutOfMemoryError} * *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code sorted} does not operate by default on a particular {@link Scheduler}.
*
* * @param sortFunction * a function that compares two items emitted by the source Publisher and returns an Integer * that indicates their sort order * @return a Flowable that emits the items emitted by the source Publisher in sorted order */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable sorted(Comparator sortFunction) { ObjectHelper.requireNonNull(sortFunction, "sortFunction"); return toList().toFlowable().map(Functions.listSorter(sortFunction)).flatMapIterable(Functions.>identity()); } /** * Returns a Flowable that emits the items in a specified {@link Iterable} before it begins to emit items * emitted by the source Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher} * is expected to honor backpressure as well. If it violates this rule, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code startWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param items * an Iterable that contains the items you want the modified Publisher to emit first * @return a Flowable that emits the items in the specified {@link Iterable} and then emits the items * emitted by the source Publisher * @see ReactiveX operators documentation: StartWith */ @SuppressWarnings("unchecked") @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable startWith(Iterable items) { return concatArray(fromIterable(items), this); } /** * Returns a Flowable that emits the items in a specified {@link Publisher} before it begins to emit * items emitted by the source Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. Both this and the {@code other} {@code Publisher}s * are expected to honor backpressure as well. If any of then violates this rule, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code startWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param other * a Publisher that contains the items you want the modified Publisher to emit first * @return a Flowable that emits the items in the specified {@link Publisher} and then emits the items * emitted by the source Publisher * @see ReactiveX operators documentation: StartWith */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable startWith(Publisher other) { ObjectHelper.requireNonNull(other, "other is null"); return concatArray(other, this); } /** * Returns a Flowable that emits a specified item before it begins to emit items emitted by the source * Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher} * is expected to honor backpressure as well. If it violates this rule, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code startWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param value * the item to emit first * @return a Flowable that emits the specified item before it begins to emit items emitted by the source * Publisher * @see ReactiveX operators documentation: StartWith */ @SuppressWarnings("unchecked") @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable startWith(T value) { ObjectHelper.requireNonNull(value, "value is null"); return concatArray(just(value), this); } /** * Returns a Flowable that emits the specified items before it begins to emit items emitted by the source * Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The source {@code Publisher} * is expected to honor backpressure as well. If it violates this rule, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code startWithArray} does not operate by default on a particular {@link Scheduler}.
*
* * @param items * the array of values to emit first * @return a Flowable that emits the specified items before it begins to emit items emitted by the source * Publisher * @see ReactiveX operators documentation: StartWith */ @SuppressWarnings("unchecked") @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable startWithArray(T... items) { Flowable fromArray = fromArray(items); if (fromArray == empty()) { return RxJavaPlugins.onAssembly(this); } return concatArray(fromArray, this); } /** * Subscribes to a Publisher and ignores {@code onNext} and {@code onComplete} emissions. *

* If the Flowable emits an error, it is wrapped into an * {@link io.reactivex.exceptions.OnErrorNotImplementedException OnErrorNotImplementedException} * and routed to the RxJavaPlugins.onError handler. *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no * backpressure is applied to it).
*
Scheduler:
*
{@code subscribe} does not operate by default on a particular {@link Scheduler}.
*
* * @return a {@link Disposable} reference with which the caller can stop receiving items before * the Publisher has finished sending them * @see ReactiveX operators documentation: Subscribe */ @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Disposable subscribe() { return subscribe(Functions.emptyConsumer(), Functions.ON_ERROR_MISSING, Functions.EMPTY_ACTION, FlowableInternalHelper.RequestMax.INSTANCE); } /** * Subscribes to a Publisher and provides a callback to handle the items it emits. *

* If the Flowable emits an error, it is wrapped into an * {@link io.reactivex.exceptions.OnErrorNotImplementedException OnErrorNotImplementedException} * and routed to the RxJavaPlugins.onError handler. *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no * backpressure is applied to it).
*
Scheduler:
*
{@code subscribe} does not operate by default on a particular {@link Scheduler}.
*
* * @param onNext * the {@code Consumer} you have designed to accept emissions from the Publisher * @return a {@link Disposable} reference with which the caller can stop receiving items before * the Publisher has finished sending them * @throws NullPointerException * if {@code onNext} is null * @see ReactiveX operators documentation: Subscribe */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Disposable subscribe(Consumer onNext) { return subscribe(onNext, Functions.ON_ERROR_MISSING, Functions.EMPTY_ACTION, FlowableInternalHelper.RequestMax.INSTANCE); } /** * Subscribes to a Publisher and provides callbacks to handle the items it emits and any error * notification it issues. *
*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no * backpressure is applied to it).
*
Scheduler:
*
{@code subscribe} does not operate by default on a particular {@link Scheduler}.
*
* * @param onNext * the {@code Consumer} you have designed to accept emissions from the Publisher * @param onError * the {@code Consumer} you have designed to accept any error notification from the * Publisher * @return a {@link Disposable} reference with which the caller can stop receiving items before * the Publisher has finished sending them * @see ReactiveX operators documentation: Subscribe * @throws NullPointerException * if {@code onNext} is null, or * if {@code onError} is null */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Disposable subscribe(Consumer onNext, Consumer onError) { return subscribe(onNext, onError, Functions.EMPTY_ACTION, FlowableInternalHelper.RequestMax.INSTANCE); } /** * Subscribes to a Publisher and provides callbacks to handle the items it emits and any error or * completion notification it issues. *
*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no * backpressure is applied to it).
*
Scheduler:
*
{@code subscribe} does not operate by default on a particular {@link Scheduler}.
*
* * @param onNext * the {@code Consumer} you have designed to accept emissions from the Publisher * @param onError * the {@code Consumer} you have designed to accept any error notification from the * Publisher * @param onComplete * the {@code Action} you have designed to accept a completion notification from the * Publisher * @return a {@link Disposable} reference with which the caller can stop receiving items before * the Publisher has finished sending them * @throws NullPointerException * if {@code onNext} is null, or * if {@code onError} is null, or * if {@code onComplete} is null * @see ReactiveX operators documentation: Subscribe */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Disposable subscribe(Consumer onNext, Consumer onError, Action onComplete) { return subscribe(onNext, onError, onComplete, FlowableInternalHelper.RequestMax.INSTANCE); } /** * Subscribes to a Publisher and provides callbacks to handle the items it emits and any error or * completion notification it issues. *
*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no * backpressure is applied to it).
*
Scheduler:
*
{@code subscribe} does not operate by default on a particular {@link Scheduler}.
*
* * @param onNext * the {@code Consumer} you have designed to accept emissions from the Publisher * @param onError * the {@code Consumer} you have designed to accept any error notification from the * Publisher * @param onComplete * the {@code Action} you have designed to accept a completion notification from the * Publisher * @param onSubscribe * the {@code Consumer} that receives the upstream's Subscription * @return a {@link Disposable} reference with which the caller can stop receiving items before * the Publisher has finished sending them * @throws NullPointerException * if {@code onNext} is null, or * if {@code onError} is null, or * if {@code onComplete} is null, or * if {@code onSubscribe} is null * @see ReactiveX operators documentation: Subscribe */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final Disposable subscribe(Consumer onNext, Consumer onError, Action onComplete, Consumer onSubscribe) { ObjectHelper.requireNonNull(onNext, "onNext is null"); ObjectHelper.requireNonNull(onError, "onError is null"); ObjectHelper.requireNonNull(onComplete, "onComplete is null"); ObjectHelper.requireNonNull(onSubscribe, "onSubscribe is null"); LambdaSubscriber ls = new LambdaSubscriber(onNext, onError, onComplete, onSubscribe); subscribe(ls); return ls; } @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) @Override public final void subscribe(Subscriber s) { if (s instanceof FlowableSubscriber) { subscribe((FlowableSubscriber)s); } else { ObjectHelper.requireNonNull(s, "s is null"); subscribe(new StrictSubscriber(s)); } } /** * Establish a connection between this Flowable and the given FlowableSubscriber and * start streaming events based on the demand of the FlowableSubscriber. *

* This is a "factory method" and can be called multiple times, each time starting a new {@link Subscription}. *

* Each {@link Subscription} will work for only a single {@link FlowableSubscriber}. *

* If the same {@link FlowableSubscriber} instance is subscribed to multiple {@link Flowable}s and/or the * same {@link Flowable} multiple times, it must ensure the serialization over its {@code onXXX} * methods manually. *

* If the {@link Flowable} rejects the subscription attempt or otherwise fails it will signal * the error via {@link FlowableSubscriber#onError(Throwable)}. *

* This subscribe method relaxes the following Reactive Streams rules: *

    *
  • §1.3: onNext should not be called concurrently until onSubscribe returns. * FlowableSubscriber.onSubscribe should make sure a sync or async call triggered by request() is safe.
  • *
  • §2.3: onError or onComplete must not call cancel. * Calling request() or cancel() is NOP at this point.
  • *
  • §2.12: onSubscribe must be called at most once on the same instance. * FlowableSubscriber reuse is not checked and if happens, it is the responsibility of * the FlowableSubscriber to ensure proper serialization of its onXXX methods.
  • *
  • §3.9: negative requests should emit an onError(IllegalArgumentException). * Non-positive requests signal via RxJavaPlugins.onError and the stream is not affected.
  • *
*
*
Backpressure:
*
The backpressure behavior/expectation is determined by the supplied {@code FlowableSubscriber}.
*
Scheduler:
*
{@code subscribe} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.0.7 - experimental; 2.1 - beta * @param s the FlowableSubscriber that will consume signals from this Flowable * @since 2.2 */ @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final void subscribe(FlowableSubscriber s) { ObjectHelper.requireNonNull(s, "s is null"); try { Subscriber z = RxJavaPlugins.onSubscribe(this, s); ObjectHelper.requireNonNull(z, "The RxJavaPlugins.onSubscribe hook returned a null FlowableSubscriber. Please check the handler provided to RxJavaPlugins.setOnFlowableSubscribe for invalid null returns. Further reading: https://github.com/ReactiveX/RxJava/wiki/Plugins"); subscribeActual(z); } catch (NullPointerException e) { // NOPMD throw e; } catch (Throwable e) { Exceptions.throwIfFatal(e); // can't call onError because no way to know if a Subscription has been set or not // can't call onSubscribe because the call might have set a Subscription already RxJavaPlugins.onError(e); NullPointerException npe = new NullPointerException("Actually not, but can't throw other exceptions due to RS"); npe.initCause(e); throw npe; } } /** * Operator implementations (both source and intermediate) should implement this method that * performs the necessary business logic and handles the incoming {@link Subscriber}s. *

There is no need to call any of the plugin hooks on the current {@code Flowable} instance or * the {@code Subscriber}; all hooks and basic safeguards have been * applied by {@link #subscribe(Subscriber)} before this method gets called. * @param s the incoming Subscriber, never null */ protected abstract void subscribeActual(Subscriber s); /** * Subscribes a given Subscriber (subclass) to this Flowable and returns the given * Subscriber as is. *

Usage example: *


     * Flowable<Integer> source = Flowable.range(1, 10);
     * CompositeDisposable composite = new CompositeDisposable();
     *
     * ResourceSubscriber<Integer> rs = new ResourceSubscriber<>() {
     *     // ...
     * };
     *
     * composite.add(source.subscribeWith(rs));
     * 
* *
*
Backpressure:
*
The backpressure behavior/expectation is determined by the supplied {@code Subscriber}.
*
Scheduler:
*
{@code subscribeWith} does not operate by default on a particular {@link Scheduler}.
*
* @param the type of the Subscriber to use and return * @param subscriber the Subscriber (subclass) to use and return, not null * @return the input {@code subscriber} * @throws NullPointerException if {@code subscriber} is null * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final > E subscribeWith(E subscriber) { subscribe(subscriber); return subscriber; } /** * Asynchronously subscribes Subscribers to this Publisher on the specified {@link Scheduler}. *

* If there is a {@link #create(FlowableOnSubscribe, BackpressureStrategy)} type source up in the * chain, it is recommended to use {@code subscribeOn(scheduler, false)} instead * to avoid same-pool deadlock because requests may pile up behind an eager/blocking emitter. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param scheduler * the {@link Scheduler} to perform subscription actions on * @return the source Publisher modified so that its subscriptions happen on the * specified {@link Scheduler} * @see ReactiveX operators documentation: SubscribeOn * @see RxJava Threading Examples * @see #observeOn * @see #subscribeOn(Scheduler, boolean) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable subscribeOn(@NonNull Scheduler scheduler) { ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return subscribeOn(scheduler, !(this instanceof FlowableCreate)); } /** * Asynchronously subscribes Subscribers to this Publisher on the specified {@link Scheduler} * optionally reroutes requests from other threads to the same {@link Scheduler} thread. *

* If there is a {@link #create(FlowableOnSubscribe, BackpressureStrategy)} type source up in the * chain, it is recommended to have {@code requestOn} false to avoid same-pool deadlock * because requests may pile up behind an eager/blocking emitter. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
*

History: 2.1.1 - experimental * @param scheduler * the {@link Scheduler} to perform subscription actions on * @param requestOn if true, requests are rerouted to the given Scheduler as well (strong pipelining) * if false, requests coming from any thread are simply forwarded to * the upstream on the same thread (weak pipelining) * @return the source Publisher modified so that its subscriptions happen on the * specified {@link Scheduler} * @see ReactiveX operators documentation: SubscribeOn * @see RxJava Threading Examples * @see #observeOn * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable subscribeOn(@NonNull Scheduler scheduler, boolean requestOn) { ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableSubscribeOn(this, scheduler, requestOn)); } /** * Returns a Flowable that emits the items emitted by the source Publisher or the items of an alternate * Publisher if the source Publisher is empty. *

* *

*
Backpressure:
*
If the source {@code Publisher} is empty, the alternate {@code Publisher} is expected to honor backpressure. * If the source {@code Publisher} is non-empty, it is expected to honor backpressure as instead. * In either case, if violated, a {@code MissingBackpressureException} may get * signaled somewhere downstream. *
*
Scheduler:
*
{@code switchIfEmpty} does not operate by default on a particular {@link Scheduler}.
*
* * @param other * the alternate Publisher to subscribe to if the source does not emit any items * @return a Publisher that emits the items emitted by the source Publisher or the items of an * alternate Publisher if the source Publisher is empty. * @since 1.1.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable switchIfEmpty(Publisher other) { ObjectHelper.requireNonNull(other, "other is null"); return RxJavaPlugins.onAssembly(new FlowableSwitchIfEmpty(this, other)); } /** * Returns a new Publisher by applying a function that you supply to each item emitted by the source * Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted * of these Publishers. *

* The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete. * If the upstream Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an * unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor * backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException} * but the violation may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code switchMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the inner Publishers and the output * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @return a Flowable that emits the items emitted by the Publisher returned from applying {@code func} to the most recently emitted item emitted by the source Publisher * @see ReactiveX operators documentation: FlatMap * @see #switchMapDelayError(Function) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable switchMap(Function> mapper) { return switchMap(mapper, bufferSize()); } /** * Returns a new Publisher by applying a function that you supply to each item emitted by the source * Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted * of these Publishers. *

* The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete. * If the upstream Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an * unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor * backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException} * but the violation may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code switchMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the inner Publishers and the output * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @param bufferSize * the number of elements to prefetch from the current active inner Publisher * @return a Flowable that emits the items emitted by the Publisher returned from applying {@code func} to the most recently emitted item emitted by the source Publisher * @see ReactiveX operators documentation: FlatMap * @see #switchMapDelayError(Function, int) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable switchMap(Function> mapper, int bufferSize) { return switchMap0(mapper, bufferSize, false); } /** * Maps the upstream values into {@link CompletableSource}s, subscribes to the newer one while * disposing the subscription to the previous {@code CompletableSource}, thus keeping at most one * active {@code CompletableSource} running. *

* *

* Since a {@code CompletableSource} doesn't produce any items, the resulting reactive type of * this operator is a {@link Completable} that can only indicate successful completion or * a failure in any of the inner {@code CompletableSource}s or the failure of the current * {@link Flowable}. *

*
Backpressure:
*
The operator consumes the current {@link Flowable} in an unbounded manner and otherwise * does not have backpressure in its return type because no items are ever produced.
*
Scheduler:
*
{@code switchMapCompletable} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
If either this {@code Flowable} or the active {@code CompletableSource} signals an {@code onError}, * the resulting {@code Completable} is terminated immediately with that {@code Throwable}. * Use the {@link #switchMapCompletableDelayError(Function)} to delay such inner failures until * every inner {@code CompletableSource}s and the main {@code Flowable} terminates in some fashion. * If they fail concurrently, the operator may combine the {@code Throwable}s into a * {@link io.reactivex.exceptions.CompositeException CompositeException} * and signal it to the downstream instead. If any inactivated (switched out) {@code CompletableSource} * signals an {@code onError} late, the {@code Throwable}s will be signaled to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. *
*
*

History: 2.1.11 - experimental * @param mapper the function called with each upstream item and should return a * {@link CompletableSource} to be subscribed to and awaited for * (non blockingly) for its terminal event * @return the new Completable instance * @see #switchMapCompletableDelayError(Function) * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Completable switchMapCompletable(@NonNull Function mapper) { ObjectHelper.requireNonNull(mapper, "mapper is null"); return RxJavaPlugins.onAssembly(new FlowableSwitchMapCompletable(this, mapper, false)); } /** * Maps the upstream values into {@link CompletableSource}s, subscribes to the newer one while * disposing the subscription to the previous {@code CompletableSource}, thus keeping at most one * active {@code CompletableSource} running and delaying any main or inner errors until all * of them terminate. *

* *

* Since a {@code CompletableSource} doesn't produce any items, the resulting reactive type of * this operator is a {@link Completable} that can only indicate successful completion or * a failure in any of the inner {@code CompletableSource}s or the failure of the current * {@link Flowable}. *

*
Backpressure:
*
The operator consumes the current {@link Flowable} in an unbounded manner and otherwise * does not have backpressure in its return type because no items are ever produced.
*
Scheduler:
*
{@code switchMapCompletableDelayError} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
Errors of this {@code Flowable} and all the {@code CompletableSource}s, who had the chance * to run to their completion, are delayed until * all of them terminate in some fashion. At this point, if there was only one failure, the respective * {@code Throwable} is emitted to the downstream. If there was more than one failure, the * operator combines all {@code Throwable}s into a {@link io.reactivex.exceptions.CompositeException CompositeException} * and signals that to the downstream. * If any inactivated (switched out) {@code CompletableSource} * signals an {@code onError} late, the {@code Throwable}s will be signaled to the global error handler via * {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. *
*
*

History: 2.1.11 - experimental * @param mapper the function called with each upstream item and should return a * {@link CompletableSource} to be subscribed to and awaited for * (non blockingly) for its terminal event * @return the new Completable instance * @see #switchMapCompletable(Function) * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Completable switchMapCompletableDelayError(@NonNull Function mapper) { ObjectHelper.requireNonNull(mapper, "mapper is null"); return RxJavaPlugins.onAssembly(new FlowableSwitchMapCompletable(this, mapper, true)); } /** * Returns a new Publisher by applying a function that you supply to each item emitted by the source * Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted * of these Publishers and delays any error until all Publishers terminate. *

* The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete. * If the upstream Publisher signals an onError, the termination of the last inner Publisher will emit that error as is * or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an * unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor * backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException} * but the violation may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code switchMapDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the inner Publishers and the output * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @return a Flowable that emits the items emitted by the Publisher returned from applying {@code func} to the most recently emitted item emitted by the source Publisher * @see ReactiveX operators documentation: FlatMap * @see #switchMap(Function) * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable switchMapDelayError(Function> mapper) { return switchMapDelayError(mapper, bufferSize()); } /** * Returns a new Publisher by applying a function that you supply to each item emitted by the source * Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted * of these Publishers and delays any error until all Publishers terminate. *

* The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete. * If the upstream Publisher signals an onError, the termination of the last inner Publisher will emit that error as is * or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an * unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor * backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException} * but the violation may lead to {@code OutOfMemoryError} due to internal buffer bloat.
*
Scheduler:
*
{@code switchMapDelayError} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the inner Publishers and the output * @param mapper * a function that, when applied to an item emitted by the source Publisher, returns a * Publisher * @param bufferSize * the number of elements to prefetch from the current active inner Publisher * @return a Flowable that emits the items emitted by the Publisher returned from applying {@code func} to the most recently emitted item emitted by the source Publisher * @see ReactiveX operators documentation: FlatMap * @see #switchMap(Function, int) * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable switchMapDelayError(Function> mapper, int bufferSize) { return switchMap0(mapper, bufferSize, true); } Flowable switchMap0(Function> mapper, int bufferSize, boolean delayError) { ObjectHelper.requireNonNull(mapper, "mapper is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); if (this instanceof ScalarCallable) { @SuppressWarnings("unchecked") T v = ((ScalarCallable)this).call(); if (v == null) { return empty(); } return FlowableScalarXMap.scalarXMap(v, mapper); } return RxJavaPlugins.onAssembly(new FlowableSwitchMap(this, mapper, bufferSize, delayError)); } /** * Maps the upstream items into {@link MaybeSource}s and switches (subscribes) to the newer ones * while disposing the older ones (and ignoring their signals) and emits the latest success value of the current one if * available while failing immediately if this {@code Flowable} or any of the * active inner {@code MaybeSource}s fail. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The main {@code Flowable} is consumed in an * unbounded manner (i.e., without backpressure).
*
Scheduler:
*
{@code switchMapMaybe} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
This operator terminates with an {@code onError} if this {@code Flowable} or any of * the inner {@code MaybeSource}s fail while they are active. When this happens concurrently, their * individual {@code Throwable} errors may get combined and emitted as a single * {@link io.reactivex.exceptions.CompositeException CompositeException}. Otherwise, a late * (i.e., inactive or switched out) {@code onError} from this {@code Flowable} or from any of * the inner {@code MaybeSource}s will be forwarded to the global error handler via * {@link io.reactivex.plugins.RxJavaPlugins#onError(Throwable)} as * {@link io.reactivex.exceptions.UndeliverableException UndeliverableException}
*
*

History: 2.1.11 - experimental * @param the output value type * @param mapper the function called with the current upstream event and should * return a {@code MaybeSource} to replace the current active inner source * and get subscribed to. * @return the new Flowable instance * @see #switchMapMaybe(Function) * @see #switchMapMaybeDelayError(Function) * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable switchMapMaybe(@NonNull Function> mapper) { ObjectHelper.requireNonNull(mapper, "mapper is null"); return RxJavaPlugins.onAssembly(new FlowableSwitchMapMaybe(this, mapper, false)); } /** * Maps the upstream items into {@link MaybeSource}s and switches (subscribes) to the newer ones * while disposing the older ones (and ignoring their signals) and emits the latest success value of the current one if * available, delaying errors from this {@code Flowable} or the inner {@code MaybeSource}s until all terminate. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The main {@code Flowable} is consumed in an * unbounded manner (i.e., without backpressure).
*
Scheduler:
*
{@code switchMapMaybeDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the output value type * @param mapper the function called with the current upstream event and should * return a {@code MaybeSource} to replace the current active inner source * and get subscribed to. * @return the new Flowable instance * @see #switchMapMaybe(Function) * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable switchMapMaybeDelayError(@NonNull Function> mapper) { ObjectHelper.requireNonNull(mapper, "mapper is null"); return RxJavaPlugins.onAssembly(new FlowableSwitchMapMaybe(this, mapper, true)); } /** * Maps the upstream items into {@link SingleSource}s and switches (subscribes) to the newer ones * while disposing the older ones (and ignoring their signals) and emits the latest success value of the current one * while failing immediately if this {@code Flowable} or any of the * active inner {@code SingleSource}s fail. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The main {@code Flowable} is consumed in an * unbounded manner (i.e., without backpressure).
*
Scheduler:
*
{@code switchMapSingle} does not operate by default on a particular {@link Scheduler}.
*
Error handling:
*
This operator terminates with an {@code onError} if this {@code Flowable} or any of * the inner {@code SingleSource}s fail while they are active. When this happens concurrently, their * individual {@code Throwable} errors may get combined and emitted as a single * {@link io.reactivex.exceptions.CompositeException CompositeException}. Otherwise, a late * (i.e., inactive or switched out) {@code onError} from this {@code Flowable} or from any of * the inner {@code SingleSource}s will be forwarded to the global error handler via * {@link io.reactivex.plugins.RxJavaPlugins#onError(Throwable)} as * {@link io.reactivex.exceptions.UndeliverableException UndeliverableException}
*
*

History: 2.1.11 - experimental * @param the output value type * @param mapper the function called with the current upstream event and should * return a {@code SingleSource} to replace the current active inner source * and get subscribed to. * @return the new Flowable instance * @see #switchMapSingleDelayError(Function) * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable switchMapSingle(@NonNull Function> mapper) { ObjectHelper.requireNonNull(mapper, "mapper is null"); return RxJavaPlugins.onAssembly(new FlowableSwitchMapSingle(this, mapper, false)); } /** * Maps the upstream items into {@link SingleSource}s and switches (subscribes) to the newer ones * while disposing the older ones (and ignoring their signals) and emits the latest success value of the current one, * delaying errors from this {@code Flowable} or the inner {@code SingleSource}s until all terminate. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The main {@code Flowable} is consumed in an * unbounded manner (i.e., without backpressure).
*
Scheduler:
*
{@code switchMapSingleDelayError} does not operate by default on a particular {@link Scheduler}.
*
*

History: 2.1.11 - experimental * @param the output value type * @param mapper the function called with the current upstream event and should * return a {@code SingleSource} to replace the current active inner source * and get subscribed to. * @return the new Flowable instance * @see #switchMapSingle(Function) * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable switchMapSingleDelayError(@NonNull Function> mapper) { ObjectHelper.requireNonNull(mapper, "mapper is null"); return RxJavaPlugins.onAssembly(new FlowableSwitchMapSingle(this, mapper, true)); } /** * Returns a Flowable that emits only the first {@code count} items emitted by the source Publisher. If the source emits fewer than * {@code count} items then all of its items are emitted. *

* *

* This method returns a Publisher that will invoke a subscribing {@link Subscriber}'s * {@link Subscriber#onNext onNext} function a maximum of {@code count} times before invoking * {@link Subscriber#onComplete onComplete}. *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior in case the first request is smaller than the {@code count}. Otherwise, the source {@code Publisher} * is consumed in an unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
This version of {@code take} does not operate by default on a particular {@link Scheduler}.
*
* * @param count * the maximum number of items to emit * @return a Flowable that emits only the first {@code count} items emitted by the source Publisher, or * all of the items from the source Publisher if that Publisher emits fewer than {@code count} items * @see ReactiveX operators documentation: Take */ @CheckReturnValue @BackpressureSupport(BackpressureKind.SPECIAL) // may trigger UNBOUNDED_IN @SchedulerSupport(SchedulerSupport.NONE) public final Flowable take(long count) { if (count < 0) { throw new IllegalArgumentException("count >= 0 required but it was " + count); } return RxJavaPlugins.onAssembly(new FlowableTake(this, count)); } /** * Returns a Flowable that emits those items emitted by source Publisher before a specified time runs * out. *

* If time runs out before the {@code Flowable} completes normally, the {@code onComplete} event will be * signaled on the default {@code computation} {@link Scheduler}. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
This version of {@code take} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @return a Flowable that emits those items emitted by the source Publisher before the time runs out * @see ReactiveX operators documentation: Take */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable take(long time, TimeUnit unit) { return takeUntil(timer(time, unit)); } /** * Returns a Flowable that emits those items emitted by source Publisher before a specified time (on a * specified Scheduler) runs out. *

* If time runs out before the {@code Flowable} completes normally, the {@code onComplete} event will be * signaled on the provided {@link Scheduler}. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param scheduler * the Scheduler used for time source * @return a Flowable that emits those items emitted by the source Publisher before the time runs out, * according to the specified Scheduler * @see ReactiveX operators documentation: Take */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable take(long time, TimeUnit unit, Scheduler scheduler) { return takeUntil(timer(time, unit, scheduler)); } /** * Returns a Flowable that emits at most the last {@code count} items emitted by the source Publisher. If the source emits fewer than * {@code count} items then all of its items are emitted. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream if the {@code count} is non-zero; ignores * backpressure if the {@code count} is zero as it doesn't signal any values.
*
Scheduler:
*
This version of {@code takeLast} does not operate by default on a particular {@link Scheduler}.
*
* * @param count * the maximum number of items to emit from the end of the sequence of items emitted by the source * Publisher * @return a Flowable that emits at most the last {@code count} items emitted by the source Publisher * @throws IndexOutOfBoundsException * if {@code count} is less than zero * @see ReactiveX operators documentation: TakeLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable takeLast(int count) { if (count < 0) { throw new IndexOutOfBoundsException("count >= 0 required but it was " + count); } else if (count == 0) { return RxJavaPlugins.onAssembly(new FlowableIgnoreElements(this)); } else if (count == 1) { return RxJavaPlugins.onAssembly(new FlowableTakeLastOne(this)); } return RxJavaPlugins.onAssembly(new FlowableTakeLast(this, count)); } /** * Returns a Flowable that emits at most a specified number of items from the source Publisher that were * emitted in a specified window of time before the Publisher completed. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., no backpressure is applied to it).
*
Scheduler:
*
{@code takeLast} does not operate on any particular scheduler but uses the current time * from the {@code computation} {@link Scheduler}.
*
* * @param count * the maximum number of items to emit * @param time * the length of the time window * @param unit * the time unit of {@code time} * @return a Flowable that emits at most {@code count} items from the source Publisher that were emitted * in a specified window of time before the Publisher completed * @see ReactiveX operators documentation: TakeLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable takeLast(long count, long time, TimeUnit unit) { return takeLast(count, time, unit, Schedulers.computation(), false, bufferSize()); } /** * Returns a Flowable that emits at most a specified number of items from the source Publisher that were * emitted in a specified window of time before the Publisher completed, where the timing information is * provided by a given Scheduler. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., no backpressure is applied to it).
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use for tracking the current time
*
* * @param count * the maximum number of items to emit * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param scheduler * the {@link Scheduler} that provides the timestamps for the observed items * @return a Flowable that emits at most {@code count} items from the source Publisher that were emitted * in a specified window of time before the Publisher completed, where the timing information is * provided by the given {@code scheduler} * @throws IndexOutOfBoundsException * if {@code count} is less than zero * @see ReactiveX operators documentation: TakeLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable takeLast(long count, long time, TimeUnit unit, Scheduler scheduler) { return takeLast(count, time, unit, scheduler, false, bufferSize()); } /** * Returns a Flowable that emits at most a specified number of items from the source Publisher that were * emitted in a specified window of time before the Publisher completed, where the timing information is * provided by a given Scheduler. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., no backpressure is applied to it).
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use for tracking the current time
*
* * @param count * the maximum number of items to emit * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param scheduler * the {@link Scheduler} that provides the timestamps for the observed items * @param delayError * if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed * by the downstream; if false, an exception is immediately signaled and all regular elements dropped * @param bufferSize * the hint about how many elements to expect to be last * @return a Flowable that emits at most {@code count} items from the source Publisher that were emitted * in a specified window of time before the Publisher completed, where the timing information is * provided by the given {@code scheduler} * @throws IndexOutOfBoundsException * if {@code count} is less than zero * @see ReactiveX operators documentation: TakeLast */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable takeLast(long count, long time, TimeUnit unit, Scheduler scheduler, boolean delayError, int bufferSize) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); if (count < 0) { throw new IndexOutOfBoundsException("count >= 0 required but it was " + count); } return RxJavaPlugins.onAssembly(new FlowableTakeLastTimed(this, count, time, unit, scheduler, bufferSize, delayError)); } /** * Returns a Flowable that emits the items from the source Publisher that were emitted in a specified * window of time before the Publisher completed. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., no backpressure is applied to it) but note that this may * lead to {@code OutOfMemoryError} due to internal buffer bloat. * Consider using {@link #takeLast(long, long, TimeUnit)} in this case.
*
Scheduler:
*
This version of {@code takeLast} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @return a Flowable that emits the items from the source Publisher that were emitted in the window of * time before the Publisher completed specified by {@code time} * @see ReactiveX operators documentation: TakeLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable takeLast(long time, TimeUnit unit) { return takeLast(time, unit, Schedulers.computation(), false, bufferSize()); } /** * Returns a Flowable that emits the items from the source Publisher that were emitted in a specified * window of time before the Publisher completed. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., no backpressure is applied to it) but note that this may * lead to {@code OutOfMemoryError} due to internal buffer bloat. * Consider using {@link #takeLast(long, long, TimeUnit)} in this case.
*
Scheduler:
*
This version of {@code takeLast} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param delayError * if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed * by the downstream; if false, an exception is immediately signaled and all regular elements dropped * @return a Flowable that emits the items from the source Publisher that were emitted in the window of * time before the Publisher completed specified by {@code time} * @see ReactiveX operators documentation: TakeLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable takeLast(long time, TimeUnit unit, boolean delayError) { return takeLast(time, unit, Schedulers.computation(), delayError, bufferSize()); } /** * Returns a Flowable that emits the items from the source Publisher that were emitted in a specified * window of time before the Publisher completed, where the timing information is provided by a specified * Scheduler. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., no backpressure is applied to it) but note that this may * lead to {@code OutOfMemoryError} due to internal buffer bloat. * Consider using {@link #takeLast(long, long, TimeUnit, Scheduler)} in this case.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param scheduler * the Scheduler that provides the timestamps for the Observed items * @return a Flowable that emits the items from the source Publisher that were emitted in the window of * time before the Publisher completed specified by {@code time}, where the timing information is * provided by {@code scheduler} * @see ReactiveX operators documentation: TakeLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable takeLast(long time, TimeUnit unit, Scheduler scheduler) { return takeLast(time, unit, scheduler, false, bufferSize()); } /** * Returns a Flowable that emits the items from the source Publisher that were emitted in a specified * window of time before the Publisher completed, where the timing information is provided by a specified * Scheduler. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., no backpressure is applied to it) but note that this may * lead to {@code OutOfMemoryError} due to internal buffer bloat. * Consider using {@link #takeLast(long, long, TimeUnit, Scheduler)} in this case.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param scheduler * the Scheduler that provides the timestamps for the Observed items * @param delayError * if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed * by the downstream; if false, an exception is immediately signaled and all regular elements dropped * @return a Flowable that emits the items from the source Publisher that were emitted in the window of * time before the Publisher completed specified by {@code time}, where the timing information is * provided by {@code scheduler} * @see ReactiveX operators documentation: TakeLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable takeLast(long time, TimeUnit unit, Scheduler scheduler, boolean delayError) { return takeLast(time, unit, scheduler, delayError, bufferSize()); } /** * Returns a Flowable that emits the items from the source Publisher that were emitted in a specified * window of time before the Publisher completed, where the timing information is provided by a specified * Scheduler. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., no backpressure is applied to it) but note that this may * lead to {@code OutOfMemoryError} due to internal buffer bloat. * Consider using {@link #takeLast(long, long, TimeUnit, Scheduler)} in this case.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param time * the length of the time window * @param unit * the time unit of {@code time} * @param scheduler * the Scheduler that provides the timestamps for the Observed items * @param delayError * if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed * by the downstream; if false, an exception is immediately signaled and all regular elements dropped * @param bufferSize * the hint about how many elements to expect to be last * @return a Flowable that emits the items from the source Publisher that were emitted in the window of * time before the Publisher completed specified by {@code time}, where the timing information is * provided by {@code scheduler} * @see ReactiveX operators documentation: TakeLast */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable takeLast(long time, TimeUnit unit, Scheduler scheduler, boolean delayError, int bufferSize) { return takeLast(Long.MAX_VALUE, time, unit, scheduler, delayError, bufferSize); } /** * Returns a Flowable that emits items emitted by the source Publisher, checks the specified predicate * for each item, and then completes when the condition is satisfied. *

* *

* The difference between this operator and {@link #takeWhile(Predicate)} is that here, the condition is * evaluated after the item is emitted. * *

*
Backpressure:
*
The operator is a pass-through for backpressure; the backpressure behavior is determined by the upstream * source and the downstream consumer.
*
Scheduler:
*
{@code takeUntil} does not operate by default on a particular {@link Scheduler}.
*
* * @param stopPredicate * a function that evaluates an item emitted by the source Publisher and returns a Boolean * @return a Flowable that first emits items emitted by the source Publisher, checks the specified * condition after each item, and then completes when the condition is satisfied. * @see ReactiveX operators documentation: TakeUntil * @see Flowable#takeWhile(Predicate) * @since 1.1.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable takeUntil(Predicate stopPredicate) { ObjectHelper.requireNonNull(stopPredicate, "stopPredicate is null"); return RxJavaPlugins.onAssembly(new FlowableTakeUntilPredicate(this, stopPredicate)); } /** * Returns a Flowable that emits the items emitted by the source Publisher until a second Publisher * emits an item. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code takeUntil} does not operate by default on a particular {@link Scheduler}.
*
* * @param other * the Publisher whose first emitted item will cause {@code takeUntil} to stop emitting items * from the source Publisher * @param * the type of items emitted by {@code other} * @return a Flowable that emits the items emitted by the source Publisher until such time as {@code other} emits its first item * @see ReactiveX operators documentation: TakeUntil */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable takeUntil(Publisher other) { ObjectHelper.requireNonNull(other, "other is null"); return RxJavaPlugins.onAssembly(new FlowableTakeUntil(this, other)); } /** * Returns a Flowable that emits items emitted by the source Publisher so long as each item satisfied a * specified condition, and then completes as soon as this condition is not satisfied. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code takeWhile} does not operate by default on a particular {@link Scheduler}.
*
* * @param predicate * a function that evaluates an item emitted by the source Publisher and returns a Boolean * @return a Flowable that emits the items from the source Publisher so long as each item satisfies the * condition defined by {@code predicate}, then completes * @see ReactiveX operators documentation: TakeWhile * @see Flowable#takeUntil(Predicate) */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable takeWhile(Predicate predicate) { ObjectHelper.requireNonNull(predicate, "predicate is null"); return RxJavaPlugins.onAssembly(new FlowableTakeWhile(this, predicate)); } /** * Returns a Flowable that emits only the first item emitted by the source Publisher during sequential * time windows of a specified duration. *

* This differs from {@link #throttleLast} in that this only tracks the passage of time whereas * {@link #throttleLast} ticks at scheduled intervals. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
{@code throttleFirst} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param windowDuration * time to wait before emitting another item after emitting the last item * @param unit * the unit of time of {@code windowDuration} * @return a Flowable that performs the throttle operation * @see ReactiveX operators documentation: Sample * @see RxJava wiki: Backpressure */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable throttleFirst(long windowDuration, TimeUnit unit) { return throttleFirst(windowDuration, unit, Schedulers.computation()); } /** * Returns a Flowable that emits only the first item emitted by the source Publisher during sequential * time windows of a specified duration, where the windows are managed by a specified Scheduler. *

* This differs from {@link #throttleLast} in that this only tracks the passage of time whereas * {@link #throttleLast} ticks at scheduled intervals. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param skipDuration * time to wait before emitting another item after emitting the last item * @param unit * the unit of time of {@code skipDuration} * @param scheduler * the {@link Scheduler} to use internally to manage the timers that handle timeout for each * event * @return a Flowable that performs the throttle operation * @see ReactiveX operators documentation: Sample * @see RxJava wiki: Backpressure */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable throttleFirst(long skipDuration, TimeUnit unit, Scheduler scheduler) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableThrottleFirstTimed(this, skipDuration, unit, scheduler)); } /** * Returns a Flowable that emits only the last item emitted by the source Publisher during sequential * time windows of a specified duration. *

* This differs from {@link #throttleFirst} in that this ticks along at a scheduled interval whereas * {@link #throttleFirst} does not tick, it just tracks the passage of time. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
{@code throttleLast} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param intervalDuration * duration of windows within which the last item emitted by the source Publisher will be * emitted * @param unit * the unit of time of {@code intervalDuration} * @return a Flowable that performs the throttle operation * @see ReactiveX operators documentation: Sample * @see RxJava wiki: Backpressure * @see #sample(long, TimeUnit) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable throttleLast(long intervalDuration, TimeUnit unit) { return sample(intervalDuration, unit); } /** * Returns a Flowable that emits only the last item emitted by the source Publisher during sequential * time windows of a specified duration, where the duration is governed by a specified Scheduler. *

* This differs from {@link #throttleFirst} in that this ticks along at a scheduled interval whereas * {@link #throttleFirst} does not tick, it just tracks the passage of time. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param intervalDuration * duration of windows within which the last item emitted by the source Publisher will be * emitted * @param unit * the unit of time of {@code intervalDuration} * @param scheduler * the {@link Scheduler} to use internally to manage the timers that handle timeout for each * event * @return a Flowable that performs the throttle operation * @see ReactiveX operators documentation: Sample * @see RxJava wiki: Backpressure * @see #sample(long, TimeUnit, Scheduler) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable throttleLast(long intervalDuration, TimeUnit unit, Scheduler scheduler) { return sample(intervalDuration, unit, scheduler); } /** * Throttles items from the upstream {@code Flowable} by first emitting the next * item from upstream, then periodically emitting the latest item (if any) when * the specified timeout elapses between them. *

* *

* Unlike the option with {@link #throttleLatest(long, TimeUnit, boolean)}, the very last item being held back * (if any) is not emitted when the upstream completes. *

* If no items were emitted from the upstream during this timeout phase, the next * upstream item is emitted immediately and the timeout window starts from then. *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow. * If the downstream is not ready to receive items, a * {@link io.reactivex.exceptions.MissingBackpressureException MissingBackpressureException} * will be signaled.
*
Scheduler:
*
{@code throttleLatest} operates by default on the {@code computation} {@link Scheduler}.
*
*

History: 2.1.14 - experimental * @param timeout the time to wait after an item emission towards the downstream * before trying to emit the latest item from upstream again * @param unit the time unit * @return the new Flowable instance * @since 2.2 * @see #throttleLatest(long, TimeUnit, boolean) * @see #throttleLatest(long, TimeUnit, Scheduler) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable throttleLatest(long timeout, TimeUnit unit) { return throttleLatest(timeout, unit, Schedulers.computation(), false); } /** * Throttles items from the upstream {@code Flowable} by first emitting the next * item from upstream, then periodically emitting the latest item (if any) when * the specified timeout elapses between them. *

* *

* If no items were emitted from the upstream during this timeout phase, the next * upstream item is emitted immediately and the timeout window starts from then. *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow. * If the downstream is not ready to receive items, a * {@link io.reactivex.exceptions.MissingBackpressureException MissingBackpressureException} * will be signaled.
*
Scheduler:
*
{@code throttleLatest} operates by default on the {@code computation} {@link Scheduler}.
*
*

History: 2.1.14 - experimental * @param timeout the time to wait after an item emission towards the downstream * before trying to emit the latest item from upstream again * @param unit the time unit * @param emitLast If {@code true}, the very last item from the upstream will be emitted * immediately when the upstream completes, regardless if there is * a timeout window active or not. If {@code false}, the very last * upstream item is ignored and the flow terminates. * @return the new Flowable instance * @see #throttleLatest(long, TimeUnit, Scheduler, boolean) * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable throttleLatest(long timeout, TimeUnit unit, boolean emitLast) { return throttleLatest(timeout, unit, Schedulers.computation(), emitLast); } /** * Throttles items from the upstream {@code Flowable} by first emitting the next * item from upstream, then periodically emitting the latest item (if any) when * the specified timeout elapses between them. *

* *

* Unlike the option with {@link #throttleLatest(long, TimeUnit, Scheduler, boolean)}, the very last item being held back * (if any) is not emitted when the upstream completes. *

* If no items were emitted from the upstream during this timeout phase, the next * upstream item is emitted immediately and the timeout window starts from then. *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow. * If the downstream is not ready to receive items, a * {@link io.reactivex.exceptions.MissingBackpressureException MissingBackpressureException} * will be signaled.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
*

History: 2.1.14 - experimental * @param timeout the time to wait after an item emission towards the downstream * before trying to emit the latest item from upstream again * @param unit the time unit * @param scheduler the {@link Scheduler} where the timed wait and latest item * emission will be performed * @return the new Flowable instance * @see #throttleLatest(long, TimeUnit, Scheduler, boolean) * @since 2.2 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable throttleLatest(long timeout, TimeUnit unit, Scheduler scheduler) { return throttleLatest(timeout, unit, scheduler, false); } /** * Throttles items from the upstream {@code Flowable} by first emitting the next * item from upstream, then periodically emitting the latest item (if any) when * the specified timeout elapses between them. *

* *

* If no items were emitted from the upstream during this timeout phase, the next * upstream item is emitted immediately and the timeout window starts from then. *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow. * If the downstream is not ready to receive items, a * {@link io.reactivex.exceptions.MissingBackpressureException MissingBackpressureException} * will be signaled.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
*

History: 2.1.14 - experimental * @param timeout the time to wait after an item emission towards the downstream * before trying to emit the latest item from upstream again * @param unit the time unit * @param scheduler the {@link Scheduler} where the timed wait and latest item * emission will be performed * @param emitLast If {@code true}, the very last item from the upstream will be emitted * immediately when the upstream completes, regardless if there is * a timeout window active or not. If {@code false}, the very last * upstream item is ignored and the flow terminates. * @return the new Flowable instance * @since 2.2 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable throttleLatest(long timeout, TimeUnit unit, Scheduler scheduler, boolean emitLast) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableThrottleLatest(this, timeout, unit, scheduler, emitLast)); } /** * Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the * source Publisher that are followed by newer items before a timeout value expires. The timer resets on * each emission (alias to {@link #debounce(long, TimeUnit)}). *

* Note: If items keep being emitted by the source Publisher faster than the timeout then no items * will be emitted by the resulting Publisher. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
{@code throttleWithTimeout} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timeout * the length of the window of time that must pass after the emission of an item from the source * Publisher in which that Publisher emits no items in order for the item to be emitted by the * resulting Publisher * @param unit * the unit of time for the specified {@code timeout} * @return a Flowable that filters out items from the source Publisher that are too quickly followed by * newer items * @see ReactiveX operators documentation: Debounce * @see RxJava wiki: Backpressure * @see #debounce(long, TimeUnit) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable throttleWithTimeout(long timeout, TimeUnit unit) { return debounce(timeout, unit); } /** * Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the * source Publisher that are followed by newer items before a timeout value expires on a specified * Scheduler. The timer resets on each emission (alias to {@link #debounce(long, TimeUnit, Scheduler)}). *

* Note: If items keep being emitted by the source Publisher faster than the timeout then no items * will be emitted by the resulting Publisher. *

* *

*
Backpressure:
*
This operator does not support backpressure as it uses time to control data flow.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timeout * the length of the window of time that must pass after the emission of an item from the source * Publisher in which that Publisher emits no items in order for the item to be emitted by the * resulting Publisher * @param unit * the unit of time for the specified {@code timeout} * @param scheduler * the {@link Scheduler} to use internally to manage the timers that handle the timeout for each * item * @return a Flowable that filters out items from the source Publisher that are too quickly followed by * newer items * @see ReactiveX operators documentation: Debounce * @see RxJava wiki: Backpressure * @see #debounce(long, TimeUnit, Scheduler) */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable throttleWithTimeout(long timeout, TimeUnit unit, Scheduler scheduler) { return debounce(timeout, unit, scheduler); } /** * Returns a Flowable that emits records of the time interval between consecutive items emitted by the * source Publisher. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code timeInterval} does not operate on any particular scheduler but uses the current time * from the {@code computation} {@link Scheduler}.
*
* * @return a Flowable that emits time interval information items * @see ReactiveX operators documentation: TimeInterval */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> timeInterval() { return timeInterval(TimeUnit.MILLISECONDS, Schedulers.computation()); } /** * Returns a Flowable that emits records of the time interval between consecutive items emitted by the * source Publisher, where this interval is computed on a specified Scheduler. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
The operator does not operate on any particular scheduler but uses the current time * from the specified {@link Scheduler}.
*
* * @param scheduler * the {@link Scheduler} used to compute time intervals * @return a Flowable that emits time interval information items * @see ReactiveX operators documentation: TimeInterval */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) // Supplied scheduler is only used for creating timestamps. public final Flowable> timeInterval(Scheduler scheduler) { return timeInterval(TimeUnit.MILLISECONDS, scheduler); } /** * Returns a Flowable that emits records of the time interval between consecutive items emitted by the * source Publisher. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code timeInterval} does not operate on any particular scheduler but uses the current time * from the {@code computation} {@link Scheduler}.
*
* * @param unit the time unit for the current time * @return a Flowable that emits time interval information items * @see ReactiveX operators documentation: TimeInterval */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> timeInterval(TimeUnit unit) { return timeInterval(unit, Schedulers.computation()); } /** * Returns a Flowable that emits records of the time interval between consecutive items emitted by the * source Publisher, where this interval is computed on a specified Scheduler. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
The operator does not operate on any particular scheduler but uses the current time * from the specified {@link Scheduler}.
*
* * @param unit the time unit for the current time * @param scheduler * the {@link Scheduler} used to compute time intervals * @return a Flowable that emits time interval information items * @see ReactiveX operators documentation: TimeInterval */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) // Supplied scheduler is only used for creating timestamps. public final Flowable> timeInterval(TimeUnit unit, Scheduler scheduler) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableTimeInterval(this, unit, scheduler)); } /** * Returns a Flowable that mirrors the source Publisher, but notifies Subscribers of a * {@code TimeoutException} if an item emitted by the source Publisher doesn't arrive within a window of * time after the emission of the previous item, where that period of time is measured by a Publisher that * is a function of the previous item. *

* *

* Note: The arrival of the first source item is never timed out. *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
This version of {@code timeout} operates by default on the {@code immediate} {@link Scheduler}.
*
* * @param * the timeout value type (ignored) * @param itemTimeoutIndicator * a function that returns a Publisher for each item emitted by the source * Publisher and that determines the timeout window for the subsequent item * @return a Flowable that mirrors the source Publisher, but notifies Subscribers of a * {@code TimeoutException} if an item emitted by the source Publisher takes longer to arrive than * the time window defined by the selector for the previously emitted item * @see ReactiveX operators documentation: Timeout */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable timeout(Function> itemTimeoutIndicator) { return timeout0(null, itemTimeoutIndicator, null); } /** * Returns a Flowable that mirrors the source Publisher, but that switches to a fallback Publisher if * an item emitted by the source Publisher doesn't arrive within a window of time after the emission of the * previous item, where that period of time is measured by a Publisher that is a function of the previous * item. *

* *

* Note: The arrival of the first source item is never timed out. *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
This version of {@code timeout} operates by default on the {@code immediate} {@link Scheduler}.
*
* * @param * the timeout value type (ignored) * @param itemTimeoutIndicator * a function that returns a Publisher, for each item emitted by the source Publisher, that * determines the timeout window for the subsequent item * @param other * the fallback Publisher to switch to if the source Publisher times out * @return a Flowable that mirrors the source Publisher, but switches to mirroring a fallback Publisher * if an item emitted by the source Publisher takes longer to arrive than the time window defined * by the selector for the previously emitted item * @see ReactiveX operators documentation: Timeout */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable timeout(Function> itemTimeoutIndicator, Flowable other) { ObjectHelper.requireNonNull(other, "other is null"); return timeout0(null, itemTimeoutIndicator, other); } /** * Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted * item. If the next item isn't emitted within the specified timeout duration starting from its predecessor, * the resulting Publisher terminates and notifies Subscribers of a {@code TimeoutException}. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
This version of {@code timeout} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timeout * maximum duration between emitted items before a timeout occurs * @param timeUnit * the unit of time that applies to the {@code timeout} argument. * @return the source Publisher modified to notify Subscribers of a {@code TimeoutException} in case of a * timeout * @see ReactiveX operators documentation: Timeout */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable timeout(long timeout, TimeUnit timeUnit) { return timeout0(timeout, timeUnit, null, Schedulers.computation()); } /** * Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted * item. If the next item isn't emitted within the specified timeout duration starting from its predecessor, * the source Publisher is disposed and resulting Publisher begins instead to mirror a fallback Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
This version of {@code timeout} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timeout * maximum duration between items before a timeout occurs * @param timeUnit * the unit of time that applies to the {@code timeout} argument * @param other * the fallback Publisher to use in case of a timeout * @return the source Publisher modified to switch to the fallback Publisher in case of a timeout * @see ReactiveX operators documentation: Timeout */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable timeout(long timeout, TimeUnit timeUnit, Publisher other) { ObjectHelper.requireNonNull(other, "other is null"); return timeout0(timeout, timeUnit, other, Schedulers.computation()); } /** * Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted * item using a specified Scheduler. If the next item isn't emitted within the specified timeout duration * starting from its predecessor, the source Publisher is disposed and resulting Publisher begins * instead to mirror a fallback Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timeout * maximum duration between items before a timeout occurs * @param timeUnit * the unit of time that applies to the {@code timeout} argument * @param scheduler * the {@link Scheduler} to run the timeout timers on * @param other * the Publisher to use as the fallback in case of a timeout * @return the source Publisher modified so that it will switch to the fallback Publisher in case of a * timeout * @see ReactiveX operators documentation: Timeout */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable timeout(long timeout, TimeUnit timeUnit, Scheduler scheduler, Publisher other) { ObjectHelper.requireNonNull(other, "other is null"); return timeout0(timeout, timeUnit, other, scheduler); } /** * Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted * item, where this policy is governed by a specified Scheduler. If the next item isn't emitted within the * specified timeout duration starting from its predecessor, the resulting Publisher terminates and * notifies Subscribers of a {@code TimeoutException}. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timeout * maximum duration between items before a timeout occurs * @param timeUnit * the unit of time that applies to the {@code timeout} argument * @param scheduler * the Scheduler to run the timeout timers on * @return the source Publisher modified to notify Subscribers of a {@code TimeoutException} in case of a * timeout * @see ReactiveX operators documentation: Timeout */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable timeout(long timeout, TimeUnit timeUnit, Scheduler scheduler) { return timeout0(timeout, timeUnit, null, scheduler); } /** * Returns a Flowable that mirrors the source Publisher, but notifies Subscribers of a * {@code TimeoutException} if either the first item emitted by the source Publisher or any subsequent item * doesn't arrive within time windows defined by other Publishers. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. Both this and the returned {@code Publisher}s * are expected to honor backpressure as well. If any of then violates this rule, it may throw an * {@code IllegalStateException} when the {@code Publisher} completes.
*
Scheduler:
*
{@code timeout} does not operate by default on any {@link Scheduler}.
*
* * @param * the first timeout value type (ignored) * @param * the subsequent timeout value type (ignored) * @param firstTimeoutIndicator * a function that returns a Publisher that determines the timeout window for the first source * item * @param itemTimeoutIndicator * a function that returns a Publisher for each item emitted by the source Publisher and that * determines the timeout window in which the subsequent source item must arrive in order to * continue the sequence * @return a Flowable that mirrors the source Publisher, but notifies Subscribers of a * {@code TimeoutException} if either the first item or any subsequent item doesn't arrive within * the time windows specified by the timeout selectors * @see ReactiveX operators documentation: Timeout */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable timeout(Publisher firstTimeoutIndicator, Function> itemTimeoutIndicator) { ObjectHelper.requireNonNull(firstTimeoutIndicator, "firstTimeoutIndicator is null"); return timeout0(firstTimeoutIndicator, itemTimeoutIndicator, null); } /** * Returns a Flowable that mirrors the source Publisher, but switches to a fallback Publisher if either * the first item emitted by the source Publisher or any subsequent item doesn't arrive within time windows * defined by other Publishers. *

* *

*
Backpressure:
*
The operator honors backpressure from downstream. The {@code Publisher} * sources are expected to honor backpressure as well. * If any of the source {@code Publisher}s violate this, it may throw an * {@code IllegalStateException} when the source {@code Publisher} completes.
*
Scheduler:
*
{@code timeout} does not operate by default on any {@link Scheduler}.
*
* * @param * the first timeout value type (ignored) * @param * the subsequent timeout value type (ignored) * @param firstTimeoutIndicator * a function that returns a Publisher which determines the timeout window for the first source * item * @param itemTimeoutIndicator * a function that returns a Publisher for each item emitted by the source Publisher and that * determines the timeout window in which the subsequent source item must arrive in order to * continue the sequence * @param other * the fallback Publisher to switch to if the source Publisher times out * @return a Flowable that mirrors the source Publisher, but switches to the {@code other} Publisher if * either the first item emitted by the source Publisher or any subsequent item doesn't arrive * within time windows defined by the timeout selectors * @throws NullPointerException * if {@code itemTimeoutIndicator} is null * @see ReactiveX operators documentation: Timeout */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable timeout( Publisher firstTimeoutIndicator, Function> itemTimeoutIndicator, Publisher other) { ObjectHelper.requireNonNull(firstTimeoutIndicator, "firstTimeoutSelector is null"); ObjectHelper.requireNonNull(other, "other is null"); return timeout0(firstTimeoutIndicator, itemTimeoutIndicator, other); } private Flowable timeout0(long timeout, TimeUnit timeUnit, Publisher other, Scheduler scheduler) { ObjectHelper.requireNonNull(timeUnit, "timeUnit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableTimeoutTimed(this, timeout, timeUnit, scheduler, other)); } private Flowable timeout0( Publisher firstTimeoutIndicator, Function> itemTimeoutIndicator, Publisher other) { ObjectHelper.requireNonNull(itemTimeoutIndicator, "itemTimeoutIndicator is null"); return RxJavaPlugins.onAssembly(new FlowableTimeout(this, firstTimeoutIndicator, itemTimeoutIndicator, other)); } /** * Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a * {@link Timed} object. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code timestamp} does not operate on any particular scheduler but uses the current time * from the {@code computation} {@link Scheduler}.
*
* * @return a Flowable that emits timestamped items from the source Publisher * @see ReactiveX operators documentation: Timestamp */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> timestamp() { return timestamp(TimeUnit.MILLISECONDS, Schedulers.computation()); } /** * Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a * {@link Timed} object whose timestamps are provided by a specified Scheduler. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
This operator does not operate on any particular scheduler but uses the current time * from the specified {@link Scheduler}.
*
* * @param scheduler * the {@link Scheduler} to use as a time source * @return a Flowable that emits timestamped items from the source Publisher with timestamps provided by * the {@code scheduler} * @see ReactiveX operators documentation: Timestamp */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) // Supplied scheduler is only used for creating timestamps. public final Flowable> timestamp(Scheduler scheduler) { return timestamp(TimeUnit.MILLISECONDS, scheduler); } /** * Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a * {@link Timed} object. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
{@code timestamp} does not operate on any particular scheduler but uses the current time * from the {@code computation} {@link Scheduler}.
*
* * @param unit the time unit for the current time * @return a Flowable that emits timestamped items from the source Publisher * @see ReactiveX operators documentation: Timestamp */ @CheckReturnValue @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> timestamp(TimeUnit unit) { return timestamp(unit, Schedulers.computation()); } /** * Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a * {@link Timed} object whose timestamps are provided by a specified Scheduler. *

* *

*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
This operator does not operate on any particular scheduler but uses the current time * from the specified {@link Scheduler}.
*
* * @param unit the time unit for the current time * @param scheduler * the {@link Scheduler} to use as a time source * @return a Flowable that emits timestamped items from the source Publisher with timestamps provided by * the {@code scheduler} * @see ReactiveX operators documentation: Timestamp */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) // Supplied scheduler is only used for creating timestamps. public final Flowable> timestamp(final TimeUnit unit, final Scheduler scheduler) { ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return map(Functions.timestampWith(unit, scheduler)); } /** * Calls the specified converter function during assembly time and returns its resulting value. *

* This allows fluent conversion to any other type. *

*
Backpressure:
*
The backpressure behavior depends on what happens in the {@code converter} function.
*
Scheduler:
*
{@code to} does not operate by default on a particular {@link Scheduler}.
*
* @param the resulting object type * @param converter the function that receives the current Flowable instance and returns a value * @return the value returned by the function */ @CheckReturnValue @BackpressureSupport(BackpressureKind.SPECIAL) @SchedulerSupport(SchedulerSupport.NONE) public final R to(Function, R> converter) { try { return ObjectHelper.requireNonNull(converter, "converter is null").apply(this); } catch (Throwable ex) { Exceptions.throwIfFatal(ex); throw ExceptionHelper.wrapOrThrow(ex); } } /** * Returns a Single that emits a single item, a list composed of all the items emitted by the * finite upstream source Publisher. *

* *

* Normally, a Publisher that returns multiple items will do so by invoking its {@link Subscriber}'s * {@link Subscriber#onNext onNext} method for each such item. You can change this behavior, instructing the * Publisher to compose a list of all of these items and then to invoke the Subscriber's {@code onNext} * function once, passing it the entire list, by calling the Publisher's {@code toList} method prior to * calling its {@link #subscribe} method. *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toList} does not operate by default on a particular {@link Scheduler}.
*
* * @return a Single that emits a single item: a List containing all of the items emitted by the source * Publisher * @see ReactiveX operators documentation: To */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single> toList() { return RxJavaPlugins.onAssembly(new FlowableToListSingle>(this)); } /** * Returns a Single that emits a single item, a list composed of all the items emitted by the * finite source Publisher. *

* *

* Normally, a Publisher that returns multiple items will do so by invoking its {@link Subscriber}'s * {@link Subscriber#onNext onNext} method for each such item. You can change this behavior, instructing the * Publisher to compose a list of all of these items and then to invoke the Subscriber's {@code onNext} * function once, passing it the entire list, by calling the Publisher's {@code toList} method prior to * calling its {@link #subscribe} method. *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toList} does not operate by default on a particular {@link Scheduler}.
*
* * @param capacityHint * the number of elements expected from the current Flowable * @return a Flowable that emits a single item: a List containing all of the items emitted by the source * Publisher * @see ReactiveX operators documentation: To */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single> toList(final int capacityHint) { ObjectHelper.verifyPositive(capacityHint, "capacityHint"); return RxJavaPlugins.onAssembly(new FlowableToListSingle>(this, Functions.createArrayList(capacityHint))); } /** * Returns a Single that emits a single item, a list composed of all the items emitted by the * finite source Publisher. *

* *

* Normally, a Publisher that returns multiple items will do so by invoking its {@link Subscriber}'s * {@link Subscriber#onNext onNext} method for each such item. You can change this behavior, instructing the * Publisher to compose a list of all of these items and then to invoke the Subscriber's {@code onNext} * function once, passing it the entire list, by calling the Publisher's {@code toList} method prior to * calling its {@link #subscribe} method. *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated collection to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toList} does not operate by default on a particular {@link Scheduler}.
*
* * @param the subclass of a collection of Ts * @param collectionSupplier * the Callable returning the collection (for each individual Subscriber) to be filled in * @return a Single that emits a single item: a List containing all of the items emitted by the source * Publisher * @see ReactiveX operators documentation: To */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final > Single toList(Callable collectionSupplier) { ObjectHelper.requireNonNull(collectionSupplier, "collectionSupplier is null"); return RxJavaPlugins.onAssembly(new FlowableToListSingle(this, collectionSupplier)); } /** * Returns a Single that emits a single HashMap containing all items emitted by the finite source Publisher, * mapped by the keys returned by a specified {@code keySelector} function. *

* *

* If more than one source item maps to the same key, the HashMap will contain the latest of those items. *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the key type of the Map * @param keySelector * the function that extracts the key from a source item to be used in the HashMap * @return a Single that emits a single item: a HashMap containing the mapped items from the source * Publisher * @see ReactiveX operators documentation: To */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single> toMap(final Function keySelector) { ObjectHelper.requireNonNull(keySelector, "keySelector is null"); return collect(HashMapSupplier.asCallable(), Functions.toMapKeySelector(keySelector)); } /** * Returns a Single that emits a single HashMap containing values corresponding to items emitted by the * finite source Publisher, mapped by the keys returned by a specified {@code keySelector} function. *

* *

* If more than one source item maps to the same key, the HashMap will contain a single entry that * corresponds to the latest of those items. *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the key type of the Map * @param the value type of the Map * @param keySelector * the function that extracts the key from a source item to be used in the HashMap * @param valueSelector * the function that extracts the value from a source item to be used in the HashMap * @return a Single that emits a single item: a HashMap containing the mapped items from the source * Publisher * @see ReactiveX operators documentation: To */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single> toMap(final Function keySelector, final Function valueSelector) { ObjectHelper.requireNonNull(keySelector, "keySelector is null"); ObjectHelper.requireNonNull(valueSelector, "valueSelector is null"); return collect(HashMapSupplier.asCallable(), Functions.toMapKeyValueSelector(keySelector, valueSelector)); } /** * Returns a Single that emits a single Map, returned by a specified {@code mapFactory} function, that * contains keys and values extracted from the items emitted by the finite source Publisher. *

* *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toMap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the key type of the Map * @param the value type of the Map * @param keySelector * the function that extracts the key from a source item to be used in the Map * @param valueSelector * the function that extracts the value from the source items to be used as value in the Map * @param mapSupplier * the function that returns a Map instance to be used * @return a Flowable that emits a single item: a Map that contains the mapped items emitted by the * source Publisher * @see ReactiveX operators documentation: To */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single> toMap(final Function keySelector, final Function valueSelector, final Callable> mapSupplier) { ObjectHelper.requireNonNull(keySelector, "keySelector is null"); ObjectHelper.requireNonNull(valueSelector, "valueSelector is null"); return collect(mapSupplier, Functions.toMapKeyValueSelector(keySelector, valueSelector)); } /** * Returns a Single that emits a single HashMap that contains an ArrayList of items emitted by the * finite source Publisher keyed by a specified {@code keySelector} function. *

* *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
This operator does not support backpressure as by intent it is requesting and buffering everything.
*
Scheduler:
*
{@code toMultimap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the key type of the Map * @param keySelector * the function that extracts the key from the source items to be used as key in the HashMap * @return a Single that emits a single item: a HashMap that contains an ArrayList of items mapped from * the source Publisher * @see ReactiveX operators documentation: To */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single>> toMultimap(Function keySelector) { Function valueSelector = Functions.identity(); Callable>> mapSupplier = HashMapSupplier.asCallable(); Function> collectionFactory = ArrayListSupplier.asFunction(); return toMultimap(keySelector, valueSelector, mapSupplier, collectionFactory); } /** * Returns a Single that emits a single HashMap that contains an ArrayList of values extracted by a * specified {@code valueSelector} function from items emitted by the finite source Publisher, keyed by a * specified {@code keySelector} function. *

* *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toMultimap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the key type of the Map * @param the value type of the Map * @param keySelector * the function that extracts a key from the source items to be used as key in the HashMap * @param valueSelector * the function that extracts a value from the source items to be used as value in the HashMap * @return a Single that emits a single item: a HashMap that contains an ArrayList of items mapped from * the source Publisher * @see ReactiveX operators documentation: To */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single>> toMultimap(Function keySelector, Function valueSelector) { Callable>> mapSupplier = HashMapSupplier.asCallable(); Function> collectionFactory = ArrayListSupplier.asFunction(); return toMultimap(keySelector, valueSelector, mapSupplier, collectionFactory); } /** * Returns a Single that emits a single Map, returned by a specified {@code mapFactory} function, that * contains a custom collection of values, extracted by a specified {@code valueSelector} function from * items emitted by the finite source Publisher, and keyed by the {@code keySelector} function. *

* *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toMultimap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the key type of the Map * @param the value type of the Map * @param keySelector * the function that extracts a key from the source items to be used as the key in the Map * @param valueSelector * the function that extracts a value from the source items to be used as the value in the Map * @param mapSupplier * the function that returns a Map instance to be used * @param collectionFactory * the function that returns a Collection instance for a particular key to be used in the Map * @return a Single that emits a single item: a Map that contains the collection of mapped items from * the source Publisher * @see ReactiveX operators documentation: To */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single>> toMultimap( final Function keySelector, final Function valueSelector, final Callable>> mapSupplier, final Function> collectionFactory) { ObjectHelper.requireNonNull(keySelector, "keySelector is null"); ObjectHelper.requireNonNull(valueSelector, "valueSelector is null"); ObjectHelper.requireNonNull(mapSupplier, "mapSupplier is null"); ObjectHelper.requireNonNull(collectionFactory, "collectionFactory is null"); return collect(mapSupplier, Functions.toMultimapKeyValueSelector(keySelector, valueSelector, collectionFactory)); } /** * Returns a Single that emits a single Map, returned by a specified {@code mapFactory} function, that * contains an ArrayList of values, extracted by a specified {@code valueSelector} function from items * emitted by the finite source Publisher and keyed by the {@code keySelector} function. *

* *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toMultimap} does not operate by default on a particular {@link Scheduler}.
*
* * @param the key type of the Map * @param the value type of the Map * @param keySelector * the function that extracts a key from the source items to be used as the key in the Map * @param valueSelector * the function that extracts a value from the source items to be used as the value in the Map * @param mapSupplier * the function that returns a Map instance to be used * @return a Single that emits a single item: a Map that contains a list items mapped from the source * Publisher * @see ReactiveX operators documentation: To */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single>> toMultimap( Function keySelector, Function valueSelector, Callable>> mapSupplier ) { return toMultimap(keySelector, valueSelector, mapSupplier, ArrayListSupplier.asFunction()); } /** * Converts the current Flowable into a non-backpressured {@link Observable}. *
*
Backpressure:
*
Observables don't support backpressure thus the current Flowable is consumed in an unbounded * manner (by requesting Long.MAX_VALUE).
*
Scheduler:
*
{@code toObservable} does not operate by default on a particular {@link Scheduler}.
*
* @return the new Observable instance * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Observable toObservable() { return RxJavaPlugins.onAssembly(new ObservableFromPublisher(this)); } /** * Returns a Single that emits a list that contains the items emitted by the finite source Publisher, in a * sorted order. Each item emitted by the Publisher must implement {@link Comparable} with respect to all * other items in the sequence. * *

If any item emitted by this Flowable does not implement {@link Comparable} with respect to * all other items emitted by this Flowable, no items will be emitted and the * sequence is terminated with a {@link ClassCastException}. *

* *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toSortedList} does not operate by default on a particular {@link Scheduler}.
*
* @return a Single that emits a list that contains the items emitted by the source Publisher in * sorted order * @see ReactiveX operators documentation: To */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single> toSortedList() { return toSortedList(Functions.naturalComparator()); } /** * Returns a Single that emits a list that contains the items emitted by the finite source Publisher, in a * sorted order based on a specified comparison function. *

* *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toSortedList} does not operate by default on a particular {@link Scheduler}.
*
* * @param comparator * a function that compares two items emitted by the source Publisher and returns an Integer * that indicates their sort order * @return a Single that emits a list that contains the items emitted by the source Publisher in * sorted order * @see ReactiveX operators documentation: To */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single> toSortedList(final Comparator comparator) { ObjectHelper.requireNonNull(comparator, "comparator is null"); return toList().map(Functions.listSorter(comparator)); } /** * Returns a Single that emits a list that contains the items emitted by the finite source Publisher, in a * sorted order based on a specified comparison function. *

* *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toSortedList} does not operate by default on a particular {@link Scheduler}.
*
* * @param comparator * a function that compares two items emitted by the source Publisher and returns an Integer * that indicates their sort order * @param capacityHint * the initial capacity of the ArrayList used to accumulate items before sorting * @return a Single that emits a list that contains the items emitted by the source Publisher in * sorted order * @see ReactiveX operators documentation: To * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single> toSortedList(final Comparator comparator, int capacityHint) { ObjectHelper.requireNonNull(comparator, "comparator is null"); return toList(capacityHint).map(Functions.listSorter(comparator)); } /** * Returns a Flowable that emits a list that contains the items emitted by the finite source Publisher, in a * sorted order. Each item emitted by the Publisher must implement {@link Comparable} with respect to all * other items in the sequence. * *

If any item emitted by this Flowable does not implement {@link Comparable} with respect to * all other items emitted by this Flowable, no items will be emitted and the * sequence is terminated with a {@link ClassCastException}. *

* *

* Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to * be emitted. Sources that are infinite and never complete will never emit anything through this * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}. *

*
Backpressure:
*
The operator honors backpressure from downstream and consumes the source {@code Publisher} in an * unbounded manner (i.e., without applying backpressure to it).
*
Scheduler:
*
{@code toSortedList} does not operate by default on a particular {@link Scheduler}.
*
* * @param capacityHint * the initial capacity of the ArrayList used to accumulate items before sorting * @return a Flowable that emits a list that contains the items emitted by the source Publisher in * sorted order * @see ReactiveX operators documentation: To * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final Single> toSortedList(int capacityHint) { return toSortedList(Functions.naturalComparator(), capacityHint); } /** * Modifies the source Publisher so that subscribers will cancel it on a specified * {@link Scheduler}. *
*
Backpressure:
*
The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure * behavior.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param scheduler * the {@link Scheduler} to perform cancellation actions on * @return the source Publisher modified so that its cancellations happen on the specified * {@link Scheduler} * @see ReactiveX operators documentation: SubscribeOn */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable unsubscribeOn(Scheduler scheduler) { ObjectHelper.requireNonNull(scheduler, "scheduler is null"); return RxJavaPlugins.onAssembly(new FlowableUnsubscribeOn(this, scheduler)); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping windows, each containing {@code count} items. When the source * Publisher completes or encounters an error, the resulting Publisher emits the current window and * propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure of its inner and outer subscribers, however, the inner Publisher uses an * unbounded buffer that may hold at most {@code count} elements.
*
Scheduler:
*
This version of {@code window} does not operate by default on a particular {@link Scheduler}.
*
* * @param count * the maximum size of each window before it should be emitted * @return a Flowable that emits connected, non-overlapping windows, each containing at most * {@code count} items from the source Publisher * @throws IllegalArgumentException if either count is non-positive * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> window(long count) { return window(count, count, bufferSize()); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits windows every {@code skip} items, each containing no more than {@code count} items. When * the source Publisher completes or encounters an error, the resulting Publisher emits the current window * and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure of its inner and outer subscribers, however, the inner Publisher uses an * unbounded buffer that may hold at most {@code count} elements.
*
Scheduler:
*
This version of {@code window} does not operate by default on a particular {@link Scheduler}.
*
* * @param count * the maximum size of each window before it should be emitted * @param skip * how many items need to be skipped before starting a new window. Note that if {@code skip} and * {@code count} are equal this is the same operation as {@link #window(long)}. * @return a Flowable that emits windows every {@code skip} items containing at most {@code count} items * from the source Publisher * @throws IllegalArgumentException if either count or skip is non-positive * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> window(long count, long skip) { return window(count, skip, bufferSize()); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits windows every {@code skip} items, each containing no more than {@code count} items. When * the source Publisher completes or encounters an error, the resulting Publisher emits the current window * and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator honors backpressure of its inner and outer subscribers, however, the inner Publisher uses an * unbounded buffer that may hold at most {@code count} elements.
*
Scheduler:
*
This version of {@code window} does not operate by default on a particular {@link Scheduler}.
*
* * @param count * the maximum size of each window before it should be emitted * @param skip * how many items need to be skipped before starting a new window. Note that if {@code skip} and * {@code count} are equal this is the same operation as {@link #window(long)}. * @param bufferSize * the capacity hint for the buffer in the inner windows * @return a Flowable that emits windows every {@code skip} items containing at most {@code count} items * from the source Publisher * @throws IllegalArgumentException if either count or skip is non-positive * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> window(long count, long skip, int bufferSize) { ObjectHelper.verifyPositive(skip, "skip"); ObjectHelper.verifyPositive(count, "count"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableWindow(this, count, skip, bufferSize)); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher starts a new window periodically, as determined by the {@code timeskip} argument. It emits * each window after a fixed timespan, specified by the {@code timespan} argument. When the source * Publisher completes or Publisher completes or encounters an error, the resulting Publisher emits the * current window and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * time to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure but have an unbounded inner buffer that may lead to {@code OutOfMemoryError} * if left unconsumed.
*
Scheduler:
*
This version of {@code window} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timespan * the period of time each window collects items before it should be emitted * @param timeskip * the period of time after which a new window will be created * @param unit * the unit of time that applies to the {@code timespan} and {@code timeskip} arguments * @return a Flowable that emits new windows periodically as a fixed timespan elapses * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable> window(long timespan, long timeskip, TimeUnit unit) { return window(timespan, timeskip, unit, Schedulers.computation(), bufferSize()); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher starts a new window periodically, as determined by the {@code timeskip} argument. It emits * each window after a fixed timespan, specified by the {@code timespan} argument. When the source * Publisher completes or Publisher completes or encounters an error, the resulting Publisher emits the * current window and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * time to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure but have an unbounded inner buffer that may lead to {@code OutOfMemoryError} * if left unconsumed.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timespan * the period of time each window collects items before it should be emitted * @param timeskip * the period of time after which a new window will be created * @param unit * the unit of time that applies to the {@code timespan} and {@code timeskip} arguments * @param scheduler * the {@link Scheduler} to use when determining the end and start of a window * @return a Flowable that emits new windows periodically as a fixed timespan elapses * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable> window(long timespan, long timeskip, TimeUnit unit, Scheduler scheduler) { return window(timespan, timeskip, unit, scheduler, bufferSize()); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher starts a new window periodically, as determined by the {@code timeskip} argument. It emits * each window after a fixed timespan, specified by the {@code timespan} argument. When the source * Publisher completes or Publisher completes or encounters an error, the resulting Publisher emits the * current window and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * time to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure but have an unbounded inner buffer that may lead to {@code OutOfMemoryError} * if left unconsumed.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timespan * the period of time each window collects items before it should be emitted * @param timeskip * the period of time after which a new window will be created * @param unit * the unit of time that applies to the {@code timespan} and {@code timeskip} arguments * @param scheduler * the {@link Scheduler} to use when determining the end and start of a window * @param bufferSize * the capacity hint for the buffer in the inner windows * @return a Flowable that emits new windows periodically as a fixed timespan elapses * @see ReactiveX operators documentation: Window */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable> window(long timespan, long timeskip, TimeUnit unit, Scheduler scheduler, int bufferSize) { ObjectHelper.verifyPositive(bufferSize, "bufferSize"); ObjectHelper.verifyPositive(timespan, "timespan"); ObjectHelper.verifyPositive(timeskip, "timeskip"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); ObjectHelper.requireNonNull(unit, "unit is null"); return RxJavaPlugins.onAssembly(new FlowableWindowTimed(this, timespan, timeskip, unit, scheduler, Long.MAX_VALUE, bufferSize, false)); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the * {@code timespan} argument. When the source Publisher completes or encounters an error, the resulting * Publisher emits the current window and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * time to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure and may hold up to {@code count} elements at most.
*
Scheduler:
*
This version of {@code window} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timespan * the period of time each window collects items before it should be emitted and replaced with a * new window * @param unit * the unit of time that applies to the {@code timespan} argument * @return a Flowable that emits connected, non-overlapping windows representing items emitted by the * source Publisher during fixed, consecutive durations * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable> window(long timespan, TimeUnit unit) { return window(timespan, unit, Schedulers.computation(), Long.MAX_VALUE, false); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping windows, each of a fixed duration as specified by the * {@code timespan} argument or a maximum size as specified by the {@code count} argument (whichever is * reached first). When the source Publisher completes or encounters an error, the resulting Publisher * emits the current window and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * time to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure and may hold up to {@code count} elements at most.
*
Scheduler:
*
This version of {@code window} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timespan * the period of time each window collects items before it should be emitted and replaced with a * new window * @param unit * the unit of time that applies to the {@code timespan} argument * @param count * the maximum size of each window before it should be emitted * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher * that were emitted during a fixed duration of time or when the window has reached maximum capacity * (whichever occurs first) * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable> window(long timespan, TimeUnit unit, long count) { return window(timespan, unit, Schedulers.computation(), count, false); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping windows, each of a fixed duration as specified by the * {@code timespan} argument or a maximum size as specified by the {@code count} argument (whichever is * reached first). When the source Publisher completes or encounters an error, the resulting Publisher * emits the current window and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * time to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure and may hold up to {@code count} elements at most.
*
Scheduler:
*
This version of {@code window} operates by default on the {@code computation} {@link Scheduler}.
*
* * @param timespan * the period of time each window collects items before it should be emitted and replaced with a * new window * @param unit * the unit of time that applies to the {@code timespan} argument * @param count * the maximum size of each window before it should be emitted * @param restart * if true, when a window reaches the capacity limit, the timer is restarted as well * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher * that were emitted during a fixed duration of time or when the window has reached maximum capacity * (whichever occurs first) * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.COMPUTATION) public final Flowable> window(long timespan, TimeUnit unit, long count, boolean restart) { return window(timespan, unit, Schedulers.computation(), count, restart); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping windows, each of a fixed duration as specified by the * {@code timespan} argument. When the source Publisher completes or encounters an error, the resulting * Publisher emits the current window and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * time to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure but have an unbounded inner buffer that may lead to {@code OutOfMemoryError} * if left unconsumed.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timespan * the period of time each window collects items before it should be emitted and replaced with a * new window * @param unit * the unit of time which applies to the {@code timespan} argument * @param scheduler * the {@link Scheduler} to use when determining the end and start of a window * @return a Flowable that emits connected, non-overlapping windows containing items emitted by the * source Publisher within a fixed duration * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable> window(long timespan, TimeUnit unit, Scheduler scheduler) { return window(timespan, unit, scheduler, Long.MAX_VALUE, false); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the * {@code timespan} argument or a maximum size specified by the {@code count} argument (whichever is reached * first). When the source Publisher completes or encounters an error, the resulting Publisher emits the * current window and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * time to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure and may hold up to {@code count} elements at most.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timespan * the period of time each window collects items before it should be emitted and replaced with a * new window * @param unit * the unit of time which applies to the {@code timespan} argument * @param count * the maximum size of each window before it should be emitted * @param scheduler * the {@link Scheduler} to use when determining the end and start of a window * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher * that were emitted during a fixed duration of time or when the window has reached maximum capacity * (whichever occurs first) * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable> window(long timespan, TimeUnit unit, Scheduler scheduler, long count) { return window(timespan, unit, scheduler, count, false); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the * {@code timespan} argument or a maximum size specified by the {@code count} argument (whichever is reached * first). When the source Publisher completes or encounters an error, the resulting Publisher emits the * current window and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * time to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure and may hold up to {@code count} elements at most.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timespan * the period of time each window collects items before it should be emitted and replaced with a * new window * @param unit * the unit of time which applies to the {@code timespan} argument * @param count * the maximum size of each window before it should be emitted * @param scheduler * the {@link Scheduler} to use when determining the end and start of a window * @param restart * if true, when a window reaches the capacity limit, the timer is restarted as well * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher * that were emitted during a fixed duration of time or when the window has reached maximum capacity * (whichever occurs first) * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable> window(long timespan, TimeUnit unit, Scheduler scheduler, long count, boolean restart) { return window(timespan, unit, scheduler, count, restart, bufferSize()); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the * {@code timespan} argument or a maximum size specified by the {@code count} argument (whichever is reached * first). When the source Publisher completes or encounters an error, the resulting Publisher emits the * current window and propagates the notification from the source Publisher. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * time to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure and may hold up to {@code count} elements at most.
*
Scheduler:
*
You specify which {@link Scheduler} this operator will use.
*
* * @param timespan * the period of time each window collects items before it should be emitted and replaced with a * new window * @param unit * the unit of time which applies to the {@code timespan} argument * @param count * the maximum size of each window before it should be emitted * @param scheduler * the {@link Scheduler} to use when determining the end and start of a window * @param restart * if true, when a window reaches the capacity limit, the timer is restarted as well * @param bufferSize * the capacity hint for the buffer in the inner windows * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher * that were emitted during a fixed duration of time or when the window has reached maximum capacity * (whichever occurs first) * @see ReactiveX operators documentation: Window */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.CUSTOM) public final Flowable> window( long timespan, TimeUnit unit, Scheduler scheduler, long count, boolean restart, int bufferSize) { ObjectHelper.verifyPositive(bufferSize, "bufferSize"); ObjectHelper.requireNonNull(scheduler, "scheduler is null"); ObjectHelper.requireNonNull(unit, "unit is null"); ObjectHelper.verifyPositive(count, "count"); return RxJavaPlugins.onAssembly(new FlowableWindowTimed(this, timespan, timespan, unit, scheduler, count, bufferSize, restart)); } /** * Returns a Flowable that emits non-overlapping windows of items it collects from the source Publisher * where the boundary of each window is determined by the items emitted from a specified boundary-governing * Publisher. *

* *

*
Backpressure:
*
The outer Publisher of this operator does not support backpressure as it uses a {@code boundary} Publisher to control data * flow. The inner Publishers honor backpressure and buffer everything until the boundary signals the next element.
*
Scheduler:
*
This version of {@code window} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the window element type (ignored) * @param boundaryIndicator * a Publisher whose emitted items close and open windows * @return a Flowable that emits non-overlapping windows of items it collects from the source Publisher * where the boundary of each window is determined by the items emitted from the {@code boundary} * Publisher * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> window(Publisher boundaryIndicator) { return window(boundaryIndicator, bufferSize()); } /** * Returns a Flowable that emits non-overlapping windows of items it collects from the source Publisher * where the boundary of each window is determined by the items emitted from a specified boundary-governing * Publisher. *

* *

*
Backpressure:
*
The outer Publisher of this operator does not support backpressure as it uses a {@code boundary} Publisher to control data * flow. The inner Publishers honor backpressure and buffer everything until the boundary signals the next element.
*
Scheduler:
*
This version of {@code window} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the window element type (ignored) * @param boundaryIndicator * a Publisher whose emitted items close and open windows * @param bufferSize * the capacity hint for the buffer in the inner windows * @return a Flowable that emits non-overlapping windows of items it collects from the source Publisher * where the boundary of each window is determined by the items emitted from the {@code boundary} * Publisher * @see ReactiveX operators documentation: Window */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> window(Publisher boundaryIndicator, int bufferSize) { ObjectHelper.requireNonNull(boundaryIndicator, "boundaryIndicator is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableWindowBoundary(this, boundaryIndicator, bufferSize)); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits windows that contain those items emitted by the source Publisher between the time when * the {@code windowOpenings} Publisher emits an item and when the Publisher returned by * {@code closingSelector} emits an item. *

* *

*
Backpressure:
*
The outer Publisher of this operator doesn't support backpressure because the emission of new * inner Publishers are controlled by the {@code windowOpenings} Publisher. * The inner Publishers honor backpressure and buffer everything until the associated closing * Publisher signals or completes.
*
Scheduler:
*
This version of {@code window} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the window-opening Publisher * @param the element type of the window-closing Publishers * @param openingIndicator * a Publisher that, when it emits an item, causes another window to be created * @param closingIndicator * a {@link Function} that produces a Publisher for every window created. When this Publisher * emits an item, the associated window is closed and emitted * @return a Flowable that emits windows of items emitted by the source Publisher that are governed by * the specified window-governing Publishers * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> window( Publisher openingIndicator, Function> closingIndicator) { return window(openingIndicator, closingIndicator, bufferSize()); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits windows that contain those items emitted by the source Publisher between the time when * the {@code windowOpenings} Publisher emits an item and when the Publisher returned by * {@code closingSelector} emits an item. *

* *

*
Backpressure:
*
The outer Publisher of this operator doesn't support backpressure because the emission of new * inner Publishers are controlled by the {@code windowOpenings} Publisher. * The inner Publishers honor backpressure and buffer everything until the associated closing * Publisher signals or completes.
*
Scheduler:
*
This version of {@code window} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the window-opening Publisher * @param the element type of the window-closing Publishers * @param openingIndicator * a Publisher that, when it emits an item, causes another window to be created * @param closingIndicator * a {@link Function} that produces a Publisher for every window created. When this Publisher * emits an item, the associated window is closed and emitted * @param bufferSize * the capacity hint for the buffer in the inner windows * @return a Flowable that emits windows of items emitted by the source Publisher that are governed by * the specified window-governing Publishers * @see ReactiveX operators documentation: Window */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> window( Publisher openingIndicator, Function> closingIndicator, int bufferSize) { ObjectHelper.requireNonNull(openingIndicator, "openingIndicator is null"); ObjectHelper.requireNonNull(closingIndicator, "closingIndicator is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableWindowBoundarySelector(this, openingIndicator, closingIndicator, bufferSize)); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping windows. It emits the current window and opens a new one * whenever the Publisher produced by the specified {@code closingSelector} emits an item. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * the {@code closingSelector} to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure but have an unbounded inner buffer that may lead to {@code OutOfMemoryError} * if left unconsumed.
*
Scheduler:
*
This version of {@code window} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the boundary Publisher * @param boundaryIndicatorSupplier * a {@link Callable} that returns a {@code Publisher} that governs the boundary between windows. * When the source {@code Publisher} emits an item, {@code window} emits the current window and begins * a new one. * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher * whenever {@code closingSelector} emits an item * @see ReactiveX operators documentation: Window */ @CheckReturnValue @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> window(Callable> boundaryIndicatorSupplier) { return window(boundaryIndicatorSupplier, bufferSize()); } /** * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting * Publisher emits connected, non-overlapping windows. It emits the current window and opens a new one * whenever the Publisher produced by the specified {@code closingSelector} emits an item. *

* *

*
Backpressure:
*
The operator consumes the source {@code Publisher} in an unbounded manner. * The returned {@code Publisher} doesn't support backpressure as it uses * the {@code closingSelector} to control the creation of windows. The returned inner {@code Publisher}s honor * backpressure but have an unbounded inner buffer that may lead to {@code OutOfMemoryError} * if left unconsumed.
*
Scheduler:
*
This version of {@code window} does not operate by default on a particular {@link Scheduler}.
*
* * @param the element type of the boundary Publisher * @param boundaryIndicatorSupplier * a {@link Callable} that returns a {@code Publisher} that governs the boundary between windows. * When the source {@code Publisher} emits an item, {@code window} emits the current window and begins * a new one. * @param bufferSize * the capacity hint for the buffer in the inner windows * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher * whenever {@code closingSelector} emits an item * @see ReactiveX operators documentation: Window */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.ERROR) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable> window(Callable> boundaryIndicatorSupplier, int bufferSize) { ObjectHelper.requireNonNull(boundaryIndicatorSupplier, "boundaryIndicatorSupplier is null"); ObjectHelper.verifyPositive(bufferSize, "bufferSize"); return RxJavaPlugins.onAssembly(new FlowableWindowBoundarySupplier(this, boundaryIndicatorSupplier, bufferSize)); } /** * Merges the specified Publisher into this Publisher sequence by using the {@code resultSelector} * function only when the source Publisher (this instance) emits an item. *

* * *

*
Backpressure:
*
The operator is a pass-through for backpressure: the backpressure support * depends on the upstream and downstream's backpressure behavior. The other Publisher * is consumed in an unbounded fashion.
*
Scheduler:
*
This operator, by default, doesn't run any particular {@link Scheduler}.
*
* * @param the element type of the other Publisher * @param the result type of the combination * @param other * the other Publisher * @param combiner * the function to call when this Publisher emits an item and the other Publisher has already * emitted an item, to generate the item to be emitted by the resulting Publisher * @return a Flowable that merges the specified Publisher into this Publisher by using the * {@code resultSelector} function only when the source Publisher sequence (this instance) emits an * item * @since 2.0 * @see ReactiveX operators documentation: CombineLatest */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable withLatestFrom(Publisher other, BiFunction combiner) { ObjectHelper.requireNonNull(other, "other is null"); ObjectHelper.requireNonNull(combiner, "combiner is null"); return RxJavaPlugins.onAssembly(new FlowableWithLatestFrom(this, combiner, other)); } /** * Combines the value emission from this Publisher with the latest emissions from the * other Publishers via a function to produce the output item. * *

Note that this operator doesn't emit anything until all other sources have produced at * least one value. The resulting emission only happens when this Publisher emits (and * not when any of the other sources emit, unlike combineLatest). * If a source doesn't produce any value and just completes, the sequence is completed immediately. * *

*
Backpressure:
*
This operator is a pass-through for backpressure behavior between the source {@code Publisher} * and the downstream Subscriber. The other {@code Publisher}s are consumed in an unbounded manner.
*
Scheduler:
*
This operator does not operate by default on a particular {@link Scheduler}.
*
* * @param the first other source's value type * @param the second other source's value type * @param the result value type * @param source1 the first other Publisher * @param source2 the second other Publisher * @param combiner the function called with an array of values from each participating Publisher * @return the new Publisher instance * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable withLatestFrom(Publisher source1, Publisher source2, Function3 combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); Function f = Functions.toFunction(combiner); return withLatestFrom(new Publisher[] { source1, source2 }, f); } /** * Combines the value emission from this Publisher with the latest emissions from the * other Publishers via a function to produce the output item. * *

Note that this operator doesn't emit anything until all other sources have produced at * least one value. The resulting emission only happens when this Publisher emits (and * not when any of the other sources emit, unlike combineLatest). * If a source doesn't produce any value and just completes, the sequence is completed immediately. * *

*
Backpressure:
*
This operator is a pass-through for backpressure behavior between the source {@code Publisher} * and the downstream Subscriber. The other {@code Publisher}s are consumed in an unbounded manner.
*
Scheduler:
*
This operator does not operate by default on a particular {@link Scheduler}.
*
* * @param the first other source's value type * @param the second other source's value type * @param the third other source's value type * @param the result value type * @param source1 the first other Publisher * @param source2 the second other Publisher * @param source3 the third other Publisher * @param combiner the function called with an array of values from each participating Publisher * @return the new Publisher instance * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable withLatestFrom( Publisher source1, Publisher source2, Publisher source3, Function4 combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); Function f = Functions.toFunction(combiner); return withLatestFrom(new Publisher[] { source1, source2, source3 }, f); } /** * Combines the value emission from this Publisher with the latest emissions from the * other Publishers via a function to produce the output item. * *

Note that this operator doesn't emit anything until all other sources have produced at * least one value. The resulting emission only happens when this Publisher emits (and * not when any of the other sources emit, unlike combineLatest). * If a source doesn't produce any value and just completes, the sequence is completed immediately. * *

*
Backpressure:
*
This operator is a pass-through for backpressure behavior between the source {@code Publisher} * and the downstream Subscriber. The other {@code Publisher}s are consumed in an unbounded manner.
*
Scheduler:
*
This operator does not operate by default on a particular {@link Scheduler}.
*
* * @param the first other source's value type * @param the second other source's value type * @param the third other source's value type * @param the fourth other source's value type * @param the result value type * @param source1 the first other Publisher * @param source2 the second other Publisher * @param source3 the third other Publisher * @param source4 the fourth other Publisher * @param combiner the function called with an array of values from each participating Publisher * @return the new Publisher instance * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable withLatestFrom( Publisher source1, Publisher source2, Publisher source3, Publisher source4, Function5 combiner) { ObjectHelper.requireNonNull(source1, "source1 is null"); ObjectHelper.requireNonNull(source2, "source2 is null"); ObjectHelper.requireNonNull(source3, "source3 is null"); ObjectHelper.requireNonNull(source4, "source4 is null"); Function f = Functions.toFunction(combiner); return withLatestFrom(new Publisher[] { source1, source2, source3, source4 }, f); } /** * Combines the value emission from this Publisher with the latest emissions from the * other Publishers via a function to produce the output item. * *

Note that this operator doesn't emit anything until all other sources have produced at * least one value. The resulting emission only happens when this Publisher emits (and * not when any of the other sources emit, unlike combineLatest). * If a source doesn't produce any value and just completes, the sequence is completed immediately. * *

*
Backpressure:
*
This operator is a pass-through for backpressure behavior between the source {@code Publisher} * and the downstream Subscriber. The other {@code Publisher}s are consumed in an unbounded manner.
*
Scheduler:
*
This operator does not operate by default on a particular {@link Scheduler}.
*
* * @param the result value type * @param others the array of other sources * @param combiner the function called with an array of values from each participating Publisher * @return the new Publisher instance * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable withLatestFrom(Publisher[] others, Function combiner) { ObjectHelper.requireNonNull(others, "others is null"); ObjectHelper.requireNonNull(combiner, "combiner is null"); return RxJavaPlugins.onAssembly(new FlowableWithLatestFromMany(this, others, combiner)); } /** * Combines the value emission from this Publisher with the latest emissions from the * other Publishers via a function to produce the output item. * *

Note that this operator doesn't emit anything until all other sources have produced at * least one value. The resulting emission only happens when this Publisher emits (and * not when any of the other sources emit, unlike combineLatest). * If a source doesn't produce any value and just completes, the sequence is completed immediately. * *

*
Backpressure:
*
This operator is a pass-through for backpressure behavior between the source {@code Publisher} * and the downstream Subscriber. The other {@code Publisher}s are consumed in an unbounded manner.
*
Scheduler:
*
This operator does not operate by default on a particular {@link Scheduler}.
*
* * @param the result value type * @param others the iterable of other sources * @param combiner the function called with an array of values from each participating Publisher * @return the new Publisher instance * @since 2.0 */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.PASS_THROUGH) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable withLatestFrom(Iterable> others, Function combiner) { ObjectHelper.requireNonNull(others, "others is null"); ObjectHelper.requireNonNull(combiner, "combiner is null"); return RxJavaPlugins.onAssembly(new FlowableWithLatestFromMany(this, others, combiner)); } /** * Returns a Flowable that emits items that are the result of applying a specified function to pairs of * values, one each from the source Publisher and a specified Iterable sequence. *

* *

* Note that the {@code other} Iterable is evaluated as items are observed from the source Publisher; it is * not pre-consumed. This allows you to zip infinite streams on either side. *

*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zipWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items in the {@code other} Iterable * @param * the type of items emitted by the resulting Publisher * @param other * the Iterable sequence * @param zipper * a function that combines the pairs of items from the Publisher and the Iterable to generate * the items to be emitted by the resulting Publisher * @return a Flowable that pairs up values from the source Publisher and the {@code other} Iterable * sequence and emits the results of {@code zipFunction} applied to these pairs * @see ReactiveX operators documentation: Zip */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable zipWith(Iterable other, BiFunction zipper) { ObjectHelper.requireNonNull(other, "other is null"); ObjectHelper.requireNonNull(zipper, "zipper is null"); return RxJavaPlugins.onAssembly(new FlowableZipIterable(this, other, zipper)); } /** * Returns a Flowable that emits items that are the result of applying a specified function to pairs of * values, one each from the source Publisher and another specified Publisher. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

range(1, 5).doOnComplete(action1).zipWith(range(6, 5).doOnComplete(action2), (a, b) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *

* *

*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zipWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the {@code other} Publisher * @param * the type of items emitted by the resulting Publisher * @param other * the other Publisher * @param zipper * a function that combines the pairs of items from the two Publishers to generate the items to * be emitted by the resulting Publisher * @return a Flowable that pairs up values from the source Publisher and the {@code other} Publisher * and emits the results of {@code zipFunction} applied to these pairs * @see ReactiveX operators documentation: Zip */ @CheckReturnValue @NonNull @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable zipWith(Publisher other, BiFunction zipper) { ObjectHelper.requireNonNull(other, "other is null"); return zip(this, other, zipper); } /** * Returns a Flowable that emits items that are the result of applying a specified function to pairs of * values, one each from the source Publisher and another specified Publisher. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

range(1, 5).doOnComplete(action1).zipWith(range(6, 5).doOnComplete(action2), (a, b) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *

* *

*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zipWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the {@code other} Publisher * @param * the type of items emitted by the resulting Publisher * @param other * the other Publisher * @param zipper * a function that combines the pairs of items from the two Publishers to generate the items to * be emitted by the resulting Publisher * @param delayError * if true, errors from the current Flowable or the other Publisher is delayed until both terminate * @return a Flowable that pairs up values from the source Publisher and the {@code other} Publisher * and emits the results of {@code zipFunction} applied to these pairs * @see ReactiveX operators documentation: Zip * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable zipWith(Publisher other, BiFunction zipper, boolean delayError) { return zip(this, other, zipper, delayError); } /** * Returns a Flowable that emits items that are the result of applying a specified function to pairs of * values, one each from the source Publisher and another specified Publisher. *

* The operator subscribes to its sources in the order they are specified and completes eagerly if * one of the sources is shorter than the rest while canceling the other sources. Therefore, it * is possible those other sources will never be able to run to completion (and thus not calling * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if * source A completes and B has been consumed and is about to complete, the operator detects A won't * be sending further values and it will cancel B immediately. For example: *

range(1, 5).doOnComplete(action1).zipWith(range(6, 5).doOnComplete(action2), (a, b) -> a + b)
* {@code action1} will be called but {@code action2} won't. *
To work around this termination property, * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion * or cancellation. *

* *

*
Backpressure:
*
The operator expects backpressure from the sources and honors backpressure from the downstream. * (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use * one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.
*
Scheduler:
*
{@code zipWith} does not operate by default on a particular {@link Scheduler}.
*
* * @param * the type of items emitted by the {@code other} Publisher * @param * the type of items emitted by the resulting Publisher * @param other * the other Publisher * @param zipper * a function that combines the pairs of items from the two Publishers to generate the items to * be emitted by the resulting Publisher * @param bufferSize * the capacity hint for the buffer in the inner windows * @param delayError * if true, errors from the current Flowable or the other Publisher is delayed until both terminate * @return a Flowable that pairs up values from the source Publisher and the {@code other} Publisher * and emits the results of {@code zipFunction} applied to these pairs * @see ReactiveX operators documentation: Zip * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final Flowable zipWith(Publisher other, BiFunction zipper, boolean delayError, int bufferSize) { return zip(this, other, zipper, delayError, bufferSize); } // ------------------------------------------------------------------------- // Fluent test support, super handy and reduces test preparation boilerplate // ------------------------------------------------------------------------- /** * Creates a TestSubscriber that requests Long.MAX_VALUE and subscribes * it to this Flowable. *
*
Backpressure:
*
The returned TestSubscriber consumes this Flowable in an unbounded fashion.
*
Scheduler:
*
{@code test} does not operate by default on a particular {@link Scheduler}.
*
* @return the new TestSubscriber instance * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.UNBOUNDED_IN) @SchedulerSupport(SchedulerSupport.NONE) public final TestSubscriber test() { // NoPMD TestSubscriber ts = new TestSubscriber(); subscribe(ts); return ts; } /** * Creates a TestSubscriber with the given initial request amount and subscribes * it to this Flowable. *
*
Backpressure:
*
The returned TestSubscriber requests the given {@code initialRequest} amount upfront.
*
Scheduler:
*
{@code test} does not operate by default on a particular {@link Scheduler}.
*
* @param initialRequest the initial request amount, positive * @return the new TestSubscriber instance * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final TestSubscriber test(long initialRequest) { // NoPMD TestSubscriber ts = new TestSubscriber(initialRequest); subscribe(ts); return ts; } /** * Creates a TestSubscriber with the given initial request amount, * optionally cancels it before the subscription and subscribes * it to this Flowable. *
*
Backpressure:
*
The returned TestSubscriber requests the given {@code initialRequest} amount upfront.
*
Scheduler:
*
{@code test} does not operate by default on a particular {@link Scheduler}.
*
* @param initialRequest the initial request amount, positive * @param cancel should the TestSubscriber be canceled before the subscription? * @return the new TestSubscriber instance * @since 2.0 */ @CheckReturnValue @BackpressureSupport(BackpressureKind.FULL) @SchedulerSupport(SchedulerSupport.NONE) public final TestSubscriber test(long initialRequest, boolean cancel) { // NoPMD TestSubscriber ts = new TestSubscriber(initialRequest); if (cancel) { ts.cancel(); } subscribe(ts); return ts; } }