OkHttp源码分析:分发器和拦截器
0.调用流程:
主要分析异步请求
OkHttpClient okHttpClient = new OkHttpClient();
Request request = new Request.Builder()
.url(url).build();
//创建一个RealCall实例
Call call=okHttpClient.newCall(request);
//RealCall.enqueue实际就是将一个RealCall放入到任务队列中,等待合适的机会执行
call.enqueue(new Callback() {
@Override
public void onFailure(Call call, IOException e) {
}
@Override
public void onResponse(Call call, Response response) throws IOException {
//response.body().string()只能调用一次
String result=response.body().string();
}
});
okhttp.png
1.分发器
内部维护队列和线程池,完成请求调配
call.enqueue
—>RealCall#enqueue
—>Dispatcher#enqueue
//异步请求等待执行队列
private final Deque<AsyncCall> readyAsyncCalls = new ArrayDeque<>();
//异步请求正在执行队列
private final Deque<AsyncCall> runningAsyncCalls = new ArrayDeque<>();
//同步请求正在执行队列
private final Deque<RealCall> runningSyncCalls = new ArrayDeque<>();
synchronized void enqueue(AsyncCall call) {
//1.正在执行的请求<64 2.相同host的请求<5
if (runningAsyncCalls.size() < maxRequests && runningCallsForHost(call) < maxRequestsPerHost) {
runningAsyncCalls.add(call);
//把任务交给线程池执行
executorService().execute(call);
} else {
readyAsyncCalls.add(call);
}
}
正在执行的请求<64并且相同host的请求<5放running,否则放ready
AsyncCall继承NamedRunnable,NamedRunnable实现了Runnable,NamedRunnable中run()又调用了execute,子类AsyncCall中的execute会被执行:
@Override protected void execute() {
boolean signalledCallback = false;
try {
//执行请求(拦截器后面分析)
Response response = getResponseWithInterceptorChain();
//...
} catch (IOException e) {
//...
} finally {
//进入finally表示这个请求任务完成
//调用分发器的finished方法
client.dispatcher().finished(this);
}
}
我们先直接看finally中的client.dispatcher().finished(this)
//Dispatcher.java
void finished(AsyncCall call) {
finished(runningAsyncCalls, call, true);
}
private <T> void finished(Deque<T> calls, T call, boolean promoteCalls) {
int runningCallsCount;
Runnable idleCallback;
synchronized (this) {
//call执行完了,从runningAsyncCalls队列中移除
if (!calls.remove(call)) throw new AssertionError("Call wasn't in-flight!");
//上面promoteCalls传入true,进入promoteCalls方法
if (promoteCalls) promoteCalls();
runningCallsCount = runningCallsCount();
idleCallback = this.idleCallback;
}
if (runningCallsCount == 0 && idleCallback != null) {
idleCallback.run();
}
}
//从readyAsyncCalls获取任务放入runningAsyncCalls队列中执行
private void promoteCalls() {
//runningAsyncCalls最大容量已满
if (runningAsyncCalls.size() >= maxRequests) return; // Already running max capacity.
//readyAsyncCalls为空
if (readyAsyncCalls.isEmpty()) return; // No ready calls to promote.
for (Iterator<AsyncCall> i = readyAsyncCalls.iterator(); i.hasNext(); ) {
//①从readyAsyncCalls中取出请求任务
AsyncCall call = i.next();
//同host请求最多只能同时5个
if (runningCallsForHost(call) < maxRequestsPerHost) {
//②将取出的任务从readyAsyncCalls中移除
i.remove();
//③将取出的任务添加到runningAsyncCalls中
runningAsyncCalls.add(call);
//④让线程池去执行
executorService().execute(call);
}
if (runningAsyncCalls.size() >= maxRequests) return; // Reached max capacity.
}
}
okhttp分发流程.png
接着我们看下线程池
public synchronized ExecutorService executorService() {
if (executorService == null) {
//核心线程数,最大线程数,线程最大闲置时间,时间单位,线程等待队列,线程创建工厂
executorService = new ThreadPoolExecutor(0, Integer.MAX_VALUE, 60, TimeUnit.SECONDS,
new SynchronousQueue<Runnable>(), Util.threadFactory("OkHttp Dispatcher", false));
}
return executorService;
}
当一个任务通过execute(Runable)方法添加到线程池时:
- 线程数量<corePoolSize,新建线程(核心)来处理被添加的任务;
- 线程数量>=corePoolSize,新任务会被添加到等待队列,添加成功则等待空闲线程,添加失败:①线程数量<maximumPoolSize,新建任务执行新任务;②线程数量>maximumPoolSize,拒绝此服务
OkHttp里面的线程池核心线程数为0,最大线程数为Integer.MAX_VALUE,无容量队列SynchronousQueue设计的目的:分发器线程池无等待,最大并发。
2.拦截器
五大拦截器完成整个请求过程
Response response = getResponseWithInterceptorChain();
//RealCall.java
Response getResponseWithInterceptorChain() throws IOException {
// Build a full stack of interceptors.
List<Interceptor> interceptors = new ArrayList<>();
//自定义拦截器
interceptors.addAll(client.interceptors());
interceptors.add(retryAndFollowUpInterceptor);
interceptors.add(new BridgeInterceptor(client.cookieJar()));
interceptors.add(new CacheInterceptor(client.internalCache()));
interceptors.add(new ConnectInterceptor(client));
if (!forWebSocket) {
interceptors.addAll(client.networkInterceptors());
}
interceptors.add(new CallServerInterceptor(forWebSocket));
//链条对象Chain
Interceptor.Chain chain = new RealInterceptorChain(interceptors, null, null, null, 0,
originalRequest, this, eventListener, client.connectTimeoutMillis(),
client.readTimeoutMillis(), client.writeTimeoutMillis());
//执行下一个拦截器
return chain.proceed(originalRequest);
}
简单介绍下五大拦截器:
1.RetryAndFollowUpInterceptor
重试拦截器:在交出(交给下一个拦截器)之前,负责判断用户是否取消了请求;在获得了结果之后,会根据响应码判断是否需要重定向(30x),处理方式是:从头部的Location中取处url地址,生成一个新的请求交给下一级(相当于重新发起一次请求)。
2.BridgeInterceptor
桥接拦截器:负责将HTTP协议必备的请求头加入其中(如HOST、Content-Type),并添加一些默认的行为(如GZIP压缩);在获得结果后,调用Cookie接口并解析GZIP数据。
3.CacheInterceptor
缓存拦截器:在交出前,读取并判断是否使用缓存;获得结果后判断是否缓存。
4.ConnectInterceptor
连接拦截器:在交出前,负责找到或新建一个连接,并获取对应的socket流。
5.CallServerInterceptor
请求服务器拦截器:真正与服务器进行通信,向服务器发送数据,解析读取响应数据。
责任链模式 —> Request请求,经过5个拦截器,发送到最下边返回响应Response,再发送到最上边。