RxJava详解之执行原理(四)
前面几篇文章介绍了RxJava的基本使用，也说了RxJava的优缺点。下面我们就从源码的角度去分析一下RxJava的原理，以及如何进行
线程切换和导致内存泄漏的原因。
// 创建被观察者、数据源
Observable<String> observable = Observable.create(new Observable.OnSubscribe<String>() {
@Override
public void call(Subscriber<? super String> subscriber) {
// 可以看到，这里传入了一个 OnSubscribe 对象作为参数。OnSubscribe 会被存储在返回的 Observable 对象中，它的作用相当于一个计划表，当 Observable
//被订阅的时候，OnSubscribe 的 call() 方法会自动被调用，事件序列就会依照设定依次触发（对于上面的代码，就是观察者Subscriber 将会被调用三次 onNext() 和一次
// onCompleted()）。这样，由被观察者调用了观察者的回调方法，就实现了由被观察者向观察者的事件传递，即观察者模式。
Log.i("@@@", "call");
subscriber.onNext("Hello ");
subscriber.onNext("World !");
subscriber.onCompleted();
}
});
// 创建观察者
Subscriber<String> subscriber = new Subscriber<String>() {
@Override
public void onCompleted() {
Log.i("@@@", "onCompleted");
}
@Override
public void onError(Throwable e) {
Log.i("@@@", "onError");
}
@Override
public void onNext(String s) {
Log.i("@@@", "onNext : " + s);
}
};
// 关联或者叫订阅更合适。
observable.subscribe(subscriber);
Observable源码分析
先看一下Observable.create()方法:
public class Observable<T> {
final OnSubscribe<T> onSubscribe;
/**
* Creates an Observable with a Function to execute when it is subscribed to.
* <p>
* <em>Note:</em> Use {@link #create(OnSubscribe)} to create an Observable, instead of this constructor,
* unless you specifically have a need for inheritance.
*
* @param f
* {@link OnSubscribe} to be executed when {@link #subscribe(Subscriber)} is called
*/
protected Observable(OnSubscribe<T> f) {
this.onSubscribe = f;
}
public static <T> Observable<T> create(OnSubscribe<T> f) {
// 调用了上面的构造函数创建一个Observable对象，而RxJavaHooks.onCreate(f)是啥呢？ 源码我们下面会看，这里我们只要知道`RxjavaHooks.onCreate(f)`的返回值还是f就行了
return new Observable<T>(RxJavaHooks.onCreate(f));
}
static <T> Subscription subscribe(Subscriber<? super T> subscriber, Observable<T> observable) {
// validate and proceed
if (subscriber == null) {
throw new IllegalArgumentException("subscriber can not be null");
}
// 这里会使用到Observable中的成员变量onSubscribe
if (observable.onSubscribe == null) {
throw new IllegalStateException("onSubscribe function can not be null.");
}
// new Subscriber so onStart it
subscriber.onStart();
// The code below is exactly the same an unsafeSubscribe but not used because it would
// add a significant depth to already huge call stacks.
try {
// allow the hook to intercept and/or decorate
// 这里会使用到Observable中的成员变量onSubscribe
RxJavaHooks.onObservableStart(observable, observable.onSubscribe).call(subscriber);
return RxJavaHooks.onObservableReturn(subscriber);
} catch (Throwable e) {
。。。
}
}
}
需要一个OnSubscribe参数，OnSbuscribe接口继承Action1接口的它是在Observale.subscribe()方法调用时会执行的操作也就是说当调用observable.subscribe()方法时就会执行到该参数的call方法:
/**
* Invoked when Observable.subscribe is called.
* @param <T> the output value type
*/
public interface OnSubscribe<T> extends Action1<Subscriber<? super T>> {
// cover for generics insanity
}
/**
* A one-argument action.
* @param <T> the first argument type
*/
public interface Action1<T> extends Action {
void call(T t);
}
而在上面Observable.create()方法中我们看到其实是调用了Observable类的构造函数，但是在参数那里传入的是RxJavaHooks.onCreate(f):
@Experimental
public final class RxJavaHooks {
static volatile Func1<Observable.OnSubscribe, Observable.OnSubscribe> onObservableCreate;
/** Initialize with the default delegation to the original RxJavaPlugins. */
// 哎~静态代码块
static {
init();
}
static void init() {
...
initCreate()
}
static void initCreate() {
onObservableCreate = new Func1<Observable.OnSubscribe, Observable.OnSubscribe>() {
@Override
public Observable.OnSubscribe call(Observable.OnSubscribe f) {
return RxJavaPlugins.getInstance().getObservableExecutionHook().onCreate(f);
}
};
onSingleCreate = new Func1<rx.Single.OnSubscribe, rx.Single.OnSubscribe>() {
@Override
public rx.Single.OnSubscribe call(rx.Single.OnSubscribe f) {
return RxJavaPlugins.getInstance().getSingleExecutionHook().onCreate(f);
}
};
onCompletableCreate = new Func1<Completable.OnSubscribe, Completable.OnSubscribe>() {
@Override
public Completable.OnSubscribe call(Completable.OnSubscribe f) {
return RxJavaPlugins.getInstance().getCompletableExecutionHook().onCreate(f);
}
};
}
/**
* Hook to call when an Observable is created.
* @param <T> the value type
* @param onSubscribe the original OnSubscribe logic
* @return the original or replacement OnSubscribe instance
*/
@SuppressWarnings({ "rawtypes", "unchecked" })
public static <T> Observable.OnSubscribe<T> onCreate(Observable.OnSubscribe<T> onSubscribe) {
// 这里onObservableCreate只有在调用initCreate()方法才会初始化，而initCreate()又被init()方法调用，init()在静态代码块中
Func1<Observable.OnSubscribe, Observable.OnSubscribe> f = onObservableCreate;
if (f != null) {
// 所以之列返回的上面onObservableCreate创建的对OnSubscribe对象,其实这里可以简单的理解为直接返回onSubscribe参数
return f.call(onSubscribe);
}
return onSubscribe;
}
public static void clear() {
if (lockdown) {
return;
}
onError = null;
onObservableCreate = null;
...
}
}
继续看一下RxJavaPlugins.getInstance().getObservableExecutionHook().onCreate(f);:
public <T> OnSubscribe<T> onCreate(OnSubscribe<T> f) {
return f;
}
到这里我们总结一下:
Observable.create(OnSubscribe f)方法其实就是内部new了一个被观察者Observable对象，同时将参数中所传过来的new出来的OnSubscribe对象赋值给了该Observable的成员变量onSubscribe。
那一定会想知道把create(OnSubscribe f)的参数赋值给成员变量onSubscribe有什么用呢？ 当然是在Observable.subscribe()和unSubscribe()中会用到啊。
Subscriber源码分析
我们在上面的例子中只是new了一个Subscriber对象并且实现了对应的方法.
Subscriber源码并不多，一共二百多行，所以....我要删除些无用的，然后都贴上来:
/**
* Provides a mechanism for receiving push-based notifications from Observables, and permits manual
* unsubscribing from these Observables.
*/
public abstract class Subscriber<T> implements Observer<T>, Subscription {
// represents requested not set yet
private static final long NOT_SET = Long.MIN_VALUE;
// Subscription that represents a group of Subscriptions that are unsubscribed together.
// 订阅事件集，所有发送给当前Subscriber的事件都会保存在这里
private final SubscriptionList subscriptions;
private final Subscriber<?> subscriber;
/**
* Interface that establishes a request-channel between an Observable and a Subscriber and allows
* the Subscriber to request a certain amount of items from the Observable (otherwise known as
* backpressure).
*/
private Producer producer;
protected Subscriber() {
this(null, false);
}
protected Subscriber(Subscriber<?> subscriber) {
this(subscriber, true);
}
/**
* Construct a Subscriber by using another Subscriber for backpressure and
* optionally for holding the subscription list (if
* <code>shareSubscriptions</code> is <code>true</code> then when
* <code>this.add(sub)</code> is called this will in fact call
* <code>subscriber.add(sub)</code>).
* <p>
* To retain the chaining of subscribers when setting
* <code>shareSubscriptions</code> to <code>false</code>, add the created
* instance to {@code subscriber} via {@link #add}.
*
* @param subscriber
* the other Subscriber
* @param shareSubscriptions
* {@code true} to share the subscription list in {@code subscriber} with
* this instance
* @since 1.0.6
*/
protected Subscriber(Subscriber<?> subscriber, boolean shareSubscriptions) {
this.subscriber = subscriber;
this.subscriptions = shareSubscriptions && subscriber != null ? subscriber.subscriptions : new SubscriptionList();
}
/**
* Adds a {@link Subscription} to this Subscriber's list of subscriptions if this list is not marked as
* unsubscribed. If the list <em>is</em> marked as unsubscribed, {@code add} will indicate this by
* explicitly unsubscribing the new {@code Subscription} as well.
*
* @param s
* the {@code Subscription} to add
*/
public final void add(Subscription s) {
subscriptions.add(s);
}
@Override
public final void unsubscribe() {
subscriptions.unsubscribe();
}
/**
* This method is invoked when the Subscriber and Observable have been connected but the Observable has
* not yet begun to emit items or send notifications to the Subscriber. Override this method to add any
* useful initialization to your subscription, for instance to initiate backpressure.
*/
public void onStart() {
// do nothing by default
}
}
可以看到Subscriber实现了Observer和Subscription接口:
public interface Observer<T> {
void onCompleted();
void onError(Throwable e);
void onNext(T t);
}
public interface Subscription {
void unsubscribe();
boolean isUnsubscribed();
}
observable.subscribe(subscriber)源码分析
好了，关系部分出现了，我们继续看Observable类中的subscribe()方法，上面在将Observable源码的时候我们简单带过这部分，没有细说，这里仔细说一下:
public class Observable<T> {
final OnSubscribe<T> onSubscribe;
protected Observable(OnSubscribe<T> f) {
this.onSubscribe = f;
}
public static <T> Observable<T> create(OnSubscribe<T> f) {
return new Observable<T>(RxJavaHooks.onCreate(f));
}
public final Subscription subscribe(Subscriber<? super T> subscriber) {
return Observable.subscribe(subscriber, this);
}
static <T> Subscription subscribe(Subscriber<? super T> subscriber, Observable<T> observable) {
// validate and proceed
if (subscriber == null) {
throw new IllegalArgumentException("subscriber can not be null");
}
if (observable.onSubscribe == null) {
throw new IllegalStateException("onSubscribe function can not be null.");
/*
* the subscribe function can also be overridden but generally that's not the appropriate approach
* so I won't mention that in the exception
*/
}
// new Subscriber so onStart it
// 1.首先调用Subscriber.onStart()方法，上面我们知道该方法没有实现
subscriber.onStart();
/*
* See https://github.com/ReactiveX/RxJava/issues/216 for discussion on "Guideline 6.4: Protect calls
* to user code from within an Observer"
*/
// if not already wrapped
// 把当前的subscriber转换成SafeSubscriber，SafeSubscriber是一个包装类，他对一些安全性做了校验，保证Subscriber中的onComplete和onError只会有一个被执行且执行一次
// 而且一旦他们执行了后就不会再继续执行onNext方法
if (!(subscriber instanceof SafeSubscriber)) {
// assign to `observer` so we return the protected version
subscriber = new SafeSubscriber<T>(subscriber);
}
// The code below is exactly the same an unsafeSubscribe but not used because it would
// add a significant depth to already huge call stacks.
try {
// allow the hook to intercept and/or decorate
// 这里就是核心了
RxJavaHooks.onObservableStart(observable, observable.onSubscribe).call(subscriber);
return RxJavaHooks.onObservableReturn(subscriber);
// 其实上面的这两句代码完全可以简写为:
// (observable.onSubscribe).call(subscriber);
// return subscriber
} catch (Throwable e) {
// special handling for certain Throwable/Error/Exception types
Exceptions.throwIfFatal(e);
// in case the subscriber can't listen to exceptions anymore
if (subscriber.isUnsubscribed()) {
RxJavaHooks.onError(RxJavaHooks.onObservableError(e));
} else {
// if an unhandled error occurs executing the onSubscribe we will propagate it
try {
subscriber.onError(RxJavaHooks.onObservableError(e));
} catch (Throwable e2) {
Exceptions.throwIfFatal(e2);
// if this happens it means the onError itself failed (perhaps an invalid function implementation)
// so we are unable to propagate the error correctly and will just throw
RuntimeException r = new OnErrorFailedException("Error occurred attempting to subscribe [" + e.getMessage() + "] and then again while trying to pass to onError.", e2);
// TODO could the hook be the cause of the error in the on error handling.
RxJavaHooks.onObservableError(r);
// TODO why aren't we throwing the hook's return value.
throw r; // NOPMD
}
}
return Subscriptions.unsubscribed();
}
}
}
上面代码的核心是:
RxJavaHooks.onObservableStart(observable, observable.onSubscribe).call(subscriber);
return RxJavaHooks.onObservableReturn(subscriber);
而RxJavaHooks这个类我们在前面讲Observable的时候也讲到过这个类的onCreate()方法。
我们分别看下他们的源码:
public static <T> Observable.OnSubscribe<T> onObservableStart(Observable<T> instance, Observable.OnSubscribe<T> onSubscribe) {
// 这一块代码是不是似曾相识？ 前面讲的RxJavaHooks。onCreate()方法是不是也是这样实现的？ 只不过那里是onObservableCreate而这里是onObservableStart，这里可以简单的理解为返回参数onSubscribe
Func2<Observable, Observable.OnSubscribe, Observable.OnSubscribe> f = onObservableStart;
if (f != null) {
return f.call(instance, onSubscribe);
}
return onSubscribe;
}
/**
* Hook to call before the Observable.subscribe() method is about to return a Subscription.
* @param subscription the original subscription
* @return the original or alternative subscription that will be returned
*/
public static Subscription onObservableReturn(Subscription subscription) {
Func1<Subscription, Subscription> f = onObservableReturn;
if (f != null) {
return f.call(subscription);
}
return subscription;
}
static void init() {
onObservableStart = new Func2<Observable, Observable.OnSubscribe, Observable.OnSubscribe>() {
@Override
public Observable.OnSubscribe call(Observable t1, Observable.OnSubscribe t2) {
return RxJavaPlugins.getInstance().getObservableExecutionHook().onSubscribeStart(t1, t2);
}
};
onObservableReturn = new Func1<Subscription, Subscription>() {
@Override
public Subscription call(Subscription f) {
return RxJavaPlugins.getInstance().getObservableExecutionHook().onSubscribeReturn(f);
}
};
}
所以这里这两句核心代码是不就可以简写为:
(observable.onSubscribe).call(subscriber);
return subscriber
而这里observable.onSubscribe变量是谁呢？ 它就是前面我们说的在Observable.onCreate(OnSubscribe<T> f)方法中将参数赋值给的那个变量，也就是OnSubscribe接口的实现类，而我们在调用onCreate()方法的时候都会实现call()方法。
/**
* Invoked when Observable.subscribe is called.
* @param <T> the output value type
*/
public interface OnSubscribe<T> extends Action1<Subscriber<? super T>> {
// cover for generics insanity
}
Over
所以到这里我们是不是就分析完了？
// 创建被观察者、数据源
Observable<String> observable = Observable.create(new Observable.OnSubscribe<String>() {
@Override
public void call(Subscriber<? super String> subscriber) {
subscriber.onNext("Hello ");
subscriber.onNext("World !");
subscriber.onCompleted();
}
});
Subscriber<String> subscriber = new Subscriber<String>() {
@Override
public void onCompleted() {
Log.i("@@@", "onCompleted");
}
@Override
public void onError(Throwable e) {
Log.i("@@@", "onError");
}
@Override
public void onNext(String s) {
Log.i("@@@", "onNext : " + s);
}
};
observable.subscribe(subscriber);
总结一下:
- 首先我们调用
Observable.create(OnSubscribe f)创建一个观察者，同时创建一个OnSubscribe接口的实现类作为create()方法的参数，并重写该接口的call()方法，在call()方法中我们调用subscriber的onNext()、onComplete()及onError()方法。而这个call()方法中的subscriber对象就是我们后面调用observable.subscribe(subscriber);所传递的subscribe对象。 - 然后调用
new Subscriber()创建一个Subscriber类的实例。 - 最后通过
observable.subscribe(subscriber);将Observable和Subscriber对象进行绑定。来订阅我们自己创建的观察者Subscriber对象。
一旦调用subscribe()方法后就会触发执行OnSubscribe.call()。而我们上面实现的OnSubscribe.call()方法会调用Subscriber类的onNext()、onComplete()及onError()等方法。
更多内容请看下一篇文章RxJava详解(五)
