XUtils3代码详解--http

这是xutils3 源码分析的第二篇。第一篇超级传送门：xutils3详解一

官方访问网络用法：

复杂版本：

/** * 自定义实体参数类请参考: * 请求注解 {@link org.xutils.http.annotation.HttpRequest} * 请求注解处理模板接口 {@link org.xutils.http.app.ParamsBuilder} * * 需要自定义类型作为callback的泛型时, 参考: * 响应注解 {@link org.xutils.http.annotation.HttpResponse} * 响应注解处理模板接口 {@link org.xutils.http.app.ResponseParser} * * 示例: 查看 org.xutils.sample.http 包里的代码 */BaiduParams params =newBaiduParams();
params.wd ="xUtils";
// 有上传文件时使用multipart表单, 否则上传原始文件流.// params.setMultipart(true);// 上传文件方式 1// params.uploadFile = new File("/sdcard/test.txt");// 上传文件方式 2// params.addBodyParameter("uploadFile", new File("/sdcard/test.txt"));Callback.Cancelable cancelable
= x.http().get(params,
/**        * 1. callback的泛型:        * callback参数默认支持的泛型类型参见{@link org.xutils.http.loader.LoaderFactory},        * 例如: 指定泛型为File则可实现文件下载, 使用params.setSaveFilePath(path)指定文件保存的全路径.        * 默认支持断点续传(采用了文件锁和尾端校验续传文件的一致性).        * 其他常用类型可以自己在LoaderFactory中注册,        * 也可以使用{@link org.xutils.http.annotation.HttpResponse}        * 将注解HttpResponse加到自定义返回值类型上, 实现自定义ResponseParser接口来统一转换.        * 如果返回值是json形式, 那么利用第三方的json工具将十分容易定义自己的ResponseParser.        * 如示例代码{@link org.xutils.sample.http.BaiduResponse}, 可直接使用BaiduResponse作为        * callback的泛型.        *        * 2. callback的组合:        * 可以用基类或接口组合个种类的Callback, 见{@link org.xutils.common.Callback}.        * 例如:        * a. 组合使用CacheCallback将使请求检测缓存或将结果存入缓存(仅GET请求生效).        * b. 组合使用PrepareCallback的prepare方法将为callback提供一次后台执行耗时任务的机会,        * 然后将结果给onCache或onSuccess.        * c. 组合使用ProgressCallback将提供进度回调.        * ...(可参考{@link org.xutils.image.ImageLoader}        * 或 示例代码中的 {@link org.xutils.sample.download.DownloadCallback})        *        * 3. 请求过程拦截或记录日志: 参考 {@link org.xutils.http.app.RequestTracker}        *        * 4. 请求Header获取: 参考 {@link org.xutils.http.app.RequestInterceptListener}        *        * 5. 其他(线程池, 超时, 重定向, 重试, 代理等): 参考 {@link org.xutils.http.RequestParams}        *        **/newCallback.CommonCallback<String>() {
@OverridepublicvoidonSuccess(Stringresult) {
Toast.makeText(x.app(), result, Toast.LENGTH_LONG).show();
}

@OverridepublicvoidonError(Throwableex, booleanisOnCallback) {
//Toast.makeText(x.app(), ex.getMessage(), Toast.LENGTH_LONG).show();if (ex instanceofHttpException) { // 网络错误HttpException httpEx = (HttpException) ex;
int responseCode = httpEx.getCode();
String responseMsg = httpEx.getMessage();
String errorResult = httpEx.getResult();
// ...
} else { // 其他错误// ...
}
Toast.makeText(x.app(), ex.getMessage(), Toast.LENGTH_LONG).show();
}

@OverridepublicvoidonCancelled(CancelledExceptioncex) {
Toast.makeText(x.app(), "cancelled", Toast.LENGTH_LONG).show();
}

@OverridepublicvoidonFinished() {

}
});

// cancelable.cancel(); // 取消请求

简单版本：

RequestParams params =newRequestParams("https://www.baidu.com/s");
params.setSslSocketFactory(...); // 设置ssl
params.addQueryStringParameter("wd", "xUtils");
x.http().get(params, newCallback.CommonCallback<String>() {
@OverridepublicvoidonSuccess(Stringresult) {
Toast.makeText(x.app(), result, Toast.LENGTH_LONG).show();
}

@OverridepublicvoidonError(Throwableex, booleanisOnCallback) {
Toast.makeText(x.app(), ex.getMessage(), Toast.LENGTH_LONG).show();
}

@OverridepublicvoidonCancelled(CancelledExceptioncex) {
Toast.makeText(x.app(), "cancelled", Toast.LENGTH_LONG).show();
}

@OverridepublicvoidonFinished() {

}
});

带有缓存的请求示例:

BaiduParams params =newBaiduParams();
params.wd ="xUtils";
// 默认缓存存活时间, 单位:毫秒.(如果服务没有返回有效的max-age或Expires)
params.setCacheMaxAge(1000*60);
Callback.Cancelable cancelable
// 使用CacheCallback, xUtils将为该请求缓存数据.= x.http().get(params, newCallback.CacheCallback<String>() {

privateboolean hasError =false;
privateString result =null;

@OverridepublicbooleanonCache(Stringresult) {
// 得到缓存数据, 缓存过期后不会进入这个方法.// 如果服务端没有返回过期时间, 参考params.setCacheMaxAge(maxAge)方法.//// * 客户端会根据服务端返回的 header 中 max-age 或 expires 来确定本地缓存是否给 onCache 方法.//   如果服务端没有返回 max-age 或 expires, 那么缓存将一直保存, 除非这里自己定义了返回false的//   逻辑, 那么xUtils将请求新数据, 来覆盖它.//// * 如果信任该缓存返回 true, 将不再请求网络;//   返回 false 继续请求网络, 但会在请求头中加上ETag, Last-Modified等信息,//   如果服务端返回304, 则表示数据没有更新, 不继续加载数据.//this.result = result;
returnfalse; // true: 信任缓存数据, 不在发起网络请求; false不信任缓存数据.
}

@OverridepublicvoidonSuccess(Stringresult) {
// 注意: 如果服务返回304或 onCache 选择了信任缓存, 这里将不会被调用,// 但是 onFinished 总会被调用.this.result = result;
}

@OverridepublicvoidonError(Throwableex, booleanisOnCallback) {
hasError =true;
Toast.makeText(x.app(), ex.getMessage(), Toast.LENGTH_LONG).show();
if (ex instanceofHttpException) { // 网络错误HttpException httpEx = (HttpException) ex;
int responseCode = httpEx.getCode();
String responseMsg = httpEx.getMessage();
String errorResult = httpEx.getResult();
// ...
} else { // 其他错误// ...
}
}

@OverridepublicvoidonCancelled(CancelledExceptioncex) {
Toast.makeText(x.app(), "cancelled", Toast.LENGTH_LONG).show();
}

@OverridepublicvoidonFinished() {
if (!hasError && result !=null) {
// 成功获取数据Toast.makeText(x.app(), result, Toast.LENGTH_LONG).show();
}
}
});

源码查看：

首先还是入口：http相关的接口和实现类。HttpManager和HttpManagerImpl

接口很简单。定义了三个异步请求和四个同步请求的方法签名。

/**
* Created by wyouflf on 15/6/17.
* http请求接口
*/
public interface HttpManager {
/**
* 异步GET请求
*/
<T> Callback.Cancelable get(RequestParams entity, Callback.CommonCallback<T> callback);
/**
* 异步POST请求
*/
<T> Callback.Cancelable post(RequestParams entity, Callback.CommonCallback<T> callback);
/**
* 异步请求
*/
<T> Callback.Cancelable request(HttpMethod method, RequestParams entity, Callback.CommonCallback<T> callback);

/**
* 同步GET请求
<T> T getSync(RequestParams entity, Class<T> resultType) throws Throwable;
/**
* 同步POST请求
*/
<T> T postSync(RequestParams entity, Class<T> resultType) throws Throwable;

/**
* 同步请求
*/
<T> T requestSync(HttpMethod method, RequestParams entity, Class<T> resultType) throws Throwable;

/**
* 同步请求
*/
<T> T requestSync(HttpMethod method, RequestParams entity, Callback.TypedCallback<T> callback) throws Throwable;
}

先看异步的一些实现方法：

HttpManagerImpl: 看源码我们得知。异步请求最后都会调用到这个方法：request（···）。

在这个方法中发现 会new一个HttpTask 并放到线程池中运行。

在上一篇博客可以知道 x.task().start(task) 其实就是调用task的doBackground方法。

@Override
public <T> Callback.Cancelable request(HttpMethod method, RequestParams entity, Callback.CommonCallback<T> callback) {
entity.setMethod(method);
Callback.Cancelable cancelable = null;
if (callback instanceof Callback.Cancelable) {
cancelable = (Callback.Cancelable) callback;
}
HttpTask<T> task = new HttpTask<T>(entity, cancelable, callback);
return x.task().start(task);
}

我们继续跟进来查看HttpTask：

我们发现httpTask 中有哦很多成员变量。 作者代码注释的也比较清楚。

/**
* Created by wyouflf on 15/7/23.
* http 请求任务
*/
public class HttpTask<ResultType> extends AbsTask<ResultType> implements ProgressHandler {
// 请求相关
private RequestParams params;
private UriRequest request;
private RequestWorker requestWorker;
private final Executor executor;
private final Callback.CommonCallback<ResultType> callback;
// 缓存控制
private Object rawResult = null;
private final Object cacheLock = new Object();
private volatile Boolean trustCache = null;
// 扩展callback
private Callback.CacheCallback<ResultType> cacheCallback;
private Callback.PrepareCallback prepareCallback;
private Callback.ProgressCallback progressCallback;
private RequestInterceptListener requestInterceptListener;
// 文件下载线程数限制
private Type loadType;
private final static int MAX_FILE_LOAD_WORKER = 3;
private final static AtomicInteger sCurrFileLoadCount = new AtomicInteger(0);

// 文件下载任务
private static final HashMap<String, WeakReference<HttpTask<?>>>
DOWNLOAD_TASK = new HashMap<String, WeakReference<HttpTask<?>>>(1);

private static final PriorityExecutor HTTP_EXECUTOR = new PriorityExecutor(5, true);
private static final PriorityExecutor CACHE_EXECUTOR = new PriorityExecutor(5, true);
}

HttpTask构造函数中,先用了两个断言来确保params 和callback 不为空。

并赋值给成员变量。 判断是哪个callback 并赋值给想应的callback

。初始化tracker 日志跟踪系统。初始化线程池 executor.

public HttpTask(RequestParams params, Callback.Cancelable cancelHandler,
Callback.CommonCallback<ResultType> callback) {
super(cancelHandler);

assert params != null;
assert callback != null;

// set params & callback
this.params = params;
this.callback = callback;
if (callback instanceof Callback.CacheCallback) {
this.cacheCallback = (Callback.CacheCallback<ResultType>) callback;
}
if (callback instanceof Callback.PrepareCallback) {
this.prepareCallback = (Callback.PrepareCallback) callback;
}
if (callback instanceof Callback.ProgressCallback) {
this.progressCallback = (Callback.ProgressCallback<ResultType>) callback;
}
if (callback instanceof RequestInterceptListener) {
this.requestInterceptListener = (RequestInterceptListener) callback;
}

// init tracker
{
RequestTracker customTracker = params.getRequestTracker();
if (customTracker == null) {
if (callback instanceof RequestTracker) {
customTracker = (RequestTracker) callback;
} else {
customTracker = UriRequestFactory.getDefaultTracker();
}
}
if (customTracker != null) {
tracker = new RequestTrackerWrapper(customTracker);
}
}

// init executor
if (params.getExecutor() != null) {
this.executor = params.getExecutor();
} else {
if (cacheCallback != null) {
this.executor = CACHE_EXECUTOR;
} else {
this.executor = HTTP_EXECUTOR;
}
}
}

doBackground 方法： 异步请求的最主要的方法。 我把这个方法截断了。 这部分主要是做一些准备工作。比如说 初始化，缓存等。

@Override
@SuppressWarnings("unchecked")
protected ResultType doBackground() throws Throwable {

if (this.isCancelled()) {
throw new Callback.CancelledException("cancelled before request");
}

// 初始化请求参数
ResultType result = null;
resolveLoadType();
request = createNewRequest();
checkDownloadTask();
// retry 初始化
boolean retry = true;
int retryCount = 0;
Throwable exception = null;
HttpRetryHandler retryHandler = this.params.getHttpRetryHandler();
if (retryHandler == null) {
retryHandler = new HttpRetryHandler();
}
retryHandler.setMaxRetryCount(this.params.getMaxRetryCount());

if (this.isCancelled()) {
throw new Callback.CancelledException("cancelled before request");
}

// 检查缓存
Object cacheResult = null;
if (cacheCallback != null && HttpMethod.permitsCache(params.getMethod())) {
// 尝试从缓存获取结果, 并为请求头加入缓存控制参数.
try {
clearRawResult();
LogUtil.d("load cache: " + this.request.getRequestUri());
rawResult = this.request.loadResultFromCache();
} catch (Throwable ex) {
LogUtil.w("load disk cache error", ex);
}

if (this.isCancelled()) {
clearRawResult();
throw new Callback.CancelledException("cancelled before request");
}

if (rawResult != null) {
if (prepareCallback != null) {
try {
cacheResult = prepareCallback.prepare(rawResult);
} catch (Throwable ex) {
cacheResult = null;
LogUtil.w("prepare disk cache error", ex);
} finally {
clearRawResult();
}
} else {
cacheResult = rawResult;
}

if (this.isCancelled()) {
throw new Callback.CancelledException("cancelled before request");
}

if (cacheResult != null) {
// 同步等待是否信任缓存
this.update(FLAG_CACHE, cacheResult);
while (trustCache == null) {
synchronized (cacheLock) {
try {
cacheLock.wait();
} catch (Throwable ignored) {
}
}
}

// 处理完成
if (trustCache) {
return null;
}
}
}
}

if (trustCache == null) {
trustCache = false;
}

if (cacheResult == null) {
this.request.clearCacheHeader();
}

发请求的代码: 请求代码中主要就是重试机子和发送机制。字段retry 是控制重试的开关。请求正常则retry一直为false。否则异常处理代码块里编写有重试的规则。

在代码里面发送请求的是对象RequestWorker的run（）；方法。所有我们去看看RequestWorker对象。

// 发起请求
retry = true;
while (retry) {
retry = false;

try {
if (this.isCancelled()) {
throw new Callback.CancelledException("cancelled before request");
}

// 由loader发起请求, 拿到结果.
this.request.close(); // retry 前关闭上次请求

try {
clearRawResult();
// 开始请求工作
LogUtil.d("load: " + this.request.getRequestUri());
requestWorker = new RequestWorker();
if (params.isCancelFast()) {
requestWorker.start();
requestWorker.join();
} else {
requestWorker.run();
}
if (requestWorker.ex != null) {
throw requestWorker.ex;
}
rawResult = requestWorker.result;
} catch (Throwable ex) {
clearRawResult();
if (this.isCancelled()) {
throw new Callback.CancelledException("cancelled during request");
} else {
throw ex;
}
}

if (prepareCallback != null) {

if (this.isCancelled()) {
throw new Callback.CancelledException("cancelled before request");
}

try {
result = (ResultType) prepareCallback.prepare(rawResult);
} finally {
clearRawResult();
}
} else {
result = (ResultType) rawResult;
}

// 保存缓存
if (cacheCallback != null && HttpMethod.permitsCache(params.getMethod())) {
this.request.save2Cache();
}

if (this.isCancelled()) {
throw new Callback.CancelledException("cancelled after request");
}
} catch (HttpRedirectException redirectEx) {
retry = true;
LogUtil.w("Http Redirect:" + params.getUri());
} catch (Throwable ex) {
if (this.request.getResponseCode() == 304) { // disk cache is valid.
return null;
} else {
exception = ex;
if (this.isCancelled() && !(exception instanceof Callback.CancelledException)) {
exception = new Callback.CancelledException("canceled by user");
}
retry = retryHandler.retryRequest(exception, ++retryCount, this.request);
}
}

}

if (exception != null && result == null && !trustCache) {
throw exception;
}

return result;
}

RequestWorker： 是HttpTask的内部类 ，主要的功能是请求发送和加载数据线程.

内部run()方法分析：

下载另作处理

请求前回调 requestInterceptListener.beforeRequest(request);

请求 this.result = request.loadResult(); 得到结果

请求后回调 requestInterceptListener.afterRequest(request);

http重定向类的异常处理。

/**
* 请求发送和加载数据线程.
* 该线程被join到HttpTask的工作线程去执行.
* 它的主要作用是为了能强行中断请求的链接过程;
* 并辅助限制同时下载文件的线程数.
* but:
* 创建一个Thread约耗时2毫秒, 优化?
*/
private final class RequestWorker extends Thread {
/*private*/ Object result;
/*private*/ Throwable ex;

private RequestWorker() {
}

public void run() {
try {
if (File.class == loadType) {
while (sCurrFileLoadCount.get() >= MAX_FILE_LOAD_WORKER
&& !HttpTask.this.isCancelled()) {
synchronized (sCurrFileLoadCount) {
try {
sCurrFileLoadCount.wait(100);
} catch (Throwable ignored) {
}
}
}
sCurrFileLoadCount.incrementAndGet();
}

if (HttpTask.this.isCancelled()) {
throw new Callback.CancelledException("cancelled before request");
}

// intercept response
if (requestInterceptListener != null) {
requestInterceptListener.beforeRequest(request);
}

try {
this.result = request.loadResult();
} catch (Throwable ex) {
this.ex = ex;
}

// intercept response
if (requestInterceptListener != null) {
requestInterceptListener.afterRequest(request);
}

if (this.ex != null) {
throw this.ex;
}
} catch (Throwable ex) {
this.ex = ex;
if (ex instanceof HttpException) {
HttpException httpEx = (HttpException) ex;
int errorCode = httpEx.getCode();
if (errorCode == 301 || errorCode == 302) {
RedirectHandler redirectHandler = params.getRedirectHandler();
if (redirectHandler != null) {
try {
RequestParams redirectParams = redirectHandler.getRedirectParams(request);
if (redirectParams != null) {
if (redirectParams.getMethod() == null) {
redirectParams.setMethod(params.getMethod());
}
// 开始重定向请求
HttpTask.this.params = redirectParams;
HttpTask.this.request = createNewRequest();
this.ex = new HttpRedirectException(errorCode, httpEx.getMessage(), httpEx.getResult());
}
} catch (Throwable throwable) {
this.ex = ex;
}
}
}
}
} finally {
if (File.class == loadType) {
synchronized (sCurrFileLoadCount) {
sCurrFileLoadCount.decrementAndGet();
sCurrFileLoadCount.notifyAll();
}
}
}
}
}

同步的一些实现方法：

HttpManagerImpl： 最终都会内部调用到requestSync（···）的方法。

在这个方法中 我们可以发现 。作者也创建了一个HttpTask。并调用了x.task().startSync(task);的方法。这个方法我们task的实现类发现实际调用的是

TaskControllerImpl.startSync( AbsTask task)

Override
public <T> T getSync(RequestParams entity, Class<T> resultType) throws Throwable {
return requestSync(HttpMethod.GET, entity, resultType);
}

@Override
public <T> T postSync(RequestParams entity, Class<T> resultType) throws Throwable {
return requestSync(HttpMethod.POST, entity, resultType);
}

@Override
public <T> T requestSync(HttpMethod method, RequestParams entity, Class<T> resultType) throws Throwable {
DefaultSyncCallback<T> callback = new DefaultSyncCallback<T>(resultType);
return requestSync(method, entity, callback);
}

@Override
public <T> T requestSync(HttpMethod method, RequestParams entity, Callback.TypedCallback<T> callback) throws Throwable {
entity.setMethod(method);
HttpTask<T> task = new HttpTask<T>(entity, null, callback);
return x.task().startSync(task);
}

TaskControllerImpl.startSync( AbsTask task) 找到这个方法我们发现 调用的是AbsTask的四个方法。onWaiting onStarted doBackground

onSuccess 我们知道其实就是httpTask的这个四个方法。通过字面意思我们也可以知道就是 等待，开始，子线程运行和成功四个方法。

@Override
public <T> T startSync(AbsTask<T> task) throws Throwable {
T result = null;
try {
task.onWaiting();
task.onStarted();
result = task.doBackground();
task.onSuccess(result);
} catch (Callback.CancelledException cex) {
task.onCancelled(cex);
} catch (Throwable ex) {
task.onError(ex, false);
throw ex;
} finally {
task.onFinished();
}
return result;
}

HttpTask的这四个方法： 在onSuccess方法中出现一个callback。回溯到TaskControllerImpl，我们发现了这个callback。它为外部传递的。外部没有传递的化。有个默认的DefaultSyncCallback。

@Override
protected void onWaiting() {
if (tracker != null) {
tracker.onWaiting(params);
}
if (progressCallback != null) {
progressCallback.onWaiting();
}
}
@Override
protected void onStarted() {
if (tracker != null) {
tracker.onStart(params);
}
if (progressCallback != null) {
progressCallback.onStarted();
}
}
@Override
protected void onSuccess(ResultType result) {
if (tracker != null) {
tracker.onSuccess(request, result);
}
if (result != null) {
callback.onSuccess(result);
}
}

DefaultSyncCallback： 此对象传递一个泛型。此泛型就为onSuccess 中返回的结果。 需要再构建DefaultSyncCallback对象时候传递。

private class DefaultSyncCallback<T> implements Callback.TypedCallback<T> {

private final Class<T> resultType;

public DefaultSyncCallback(Class<T> resultType) {
this.resultType = resultType;
}

@Override
public Type getLoadType() {
return resultType;
}

@Override
public void onSuccess(T result) {

}

@Override
public void onError(Throwable ex, boolean isOnCallback) {

}

@Override
public void onCancelled(CancelledException cex) {

}

@Override
public void onFinished() {

}
}

从task的onwaiting 可以得到。同步请求会一直等待 直到请求返回 或者请求超时。