上一篇文章从功能和用法上对AsyncHttpClient做了个大致介绍，今天我们和以往一样，从内部实现、原理的角度带领大家看看
其工作机制，以便加深理解。写程序越多，我发现每接触一个新东西，都会有强烈的想知道它内部怎么工作的冲动。可能只有知道了
内部原理能更容易写出高质量的代码吧。
我大概浏览了下其代码，关键部分可以分为这4个模块：
1. AsyncHttpClient自己一个模块；
2. AsyncHttpRequest和RequestHandler一个模块；
3. AsyncHttpResponseHandler及其各种特定子类一个模块；
4. RetryHandler，自动重试机制。
我们可以很清楚的看出门道来，大体是按照client、request、response，这样的方式组织的。接下来我们的代码分析也就按照这个顺序进行。
先来说AsyncHttpClient，来看其关键字段和ctor，代码如下：
public static final String LOG_TAG = "AsyncHttpClient"; public static final String HEADER_CONTENT_TYPE = "Content-Type"; public static final String HEADER_CONTENT_RANGE = "Content-Range"; public static final String HEADER_CONTENT_ENCODING = "Content-Encoding"; public static final String HEADER_CONTENT_DISPOSITION = "Content-Disposition"; public static final String HEADER_ACCEPT_ENCODING = "Accept-Encoding"; public static final String ENCODING_GZIP = "gzip"; public static final int DEFAULT_MAX_CONNECTIONS = 10; public static final int DEFAULT_SOCKET_TIMEOUT = 10 * 1000; public static final int DEFAULT_MAX_RETRIES = 5; public static final int DEFAULT_RETRY_SLEEP_TIME_MILLIS = 1500; public static final int DEFAULT_SOCKET_BUFFER_SIZE = 8192; private int maxConnections = DEFAULT_MAX_CONNECTIONS; private int connectTimeout = DEFAULT_SOCKET_TIMEOUT; private int responseTimeout = DEFAULT_SOCKET_TIMEOUT; // 各种参数设置 private final DefaultHttpClient httpClient; // 包装的Apache DefaultHttpClient private final HttpContext httpContext; private ExecutorService threadPool; // 执行网络请求的线程池 private final Map<Context, List<RequestHandle>> requestMap; // 与Android Context对应的请求map private final Map<String, String> clientHeaderMap; // 客户端的请求header map private boolean isUrlEncodingEnabled = true; // 允许url encoding
接下来看看各种ctor，如下：
/** * Creates a new AsyncHttpClient with default constructor arguments values */ public AsyncHttpClient() { // 一般客户端代码中都直接调用这个版本的ctor this(false, 80, 443); } /** * Creates a new AsyncHttpClient. * * @param httpPort non-standard HTTP-only port */ public AsyncHttpClient(int httpPort) { this(false, httpPort, 443); } /** * Creates a new AsyncHttpClient. * * @param httpPort non-standard HTTP-only port * @param httpsPort non-standard HTTPS-only port */ public AsyncHttpClient(int httpPort, int httpsPort) { this(false, httpPort, httpsPort); } /** * Creates new AsyncHttpClient using given params * * @param fixNoHttpResponseException Whether to fix or not issue, by omitting SSL verification * @param httpPort HTTP port to be used, must be greater than 0 * @param httpsPort HTTPS port to be used, must be greater than 0 */ public AsyncHttpClient(boolean fixNoHttpResponseException, int httpPort, int httpsPort) { this(getDefaultSchemeRegistry(fixNoHttpResponseException, httpPort, httpsPort)); } /** * Returns default instance of SchemeRegistry * * @param fixNoHttpResponseException Whether to fix or not issue, by omitting SSL verification * @param httpPort HTTP port to be used, must be greater than 0 * @param httpsPort HTTPS port to be used, must be greater than 0 */ private static SchemeRegistry getDefaultSchemeRegistry(boolean fixNoHttpResponseException, int httpPort, int httpsPort) { if (fixNoHttpResponseException) { // 如果你请求的url是https的，并且遇到了SSL验证之类的错误，那么你应该将此值设为true试试 Log.d(LOG_TAG, "Beware! Using the fix is insecure, as it doesn‘t verify SSL certificates."); } if (httpPort < 1) { httpPort = 80; Log.d(LOG_TAG, "Invalid HTTP port number specified, defaulting to 80"); } if (httpsPort < 1) { httpsPort = 443; Log.d(LOG_TAG, "Invalid HTTPS port number specified, defaulting to 443"); } // Fix to SSL flaw in API < ICS // See https://code.google.com/p/android/issues/detail?id=13117 SSLSocketFactory sslSocketFactory; if (fixNoHttpResponseException) { // 感兴趣的同学可自行看看MySSLSocketFactory的实现，基本上是省略了SSL验证环节 sslSocketFactory = MySSLSocketFactory.getFixedSocketFactory(); } else { sslSocketFactory = SSLSocketFactory.getSocketFactory(); } SchemeRegistry schemeRegistry = new SchemeRegistry(); schemeRegistry.register(new Scheme("http", PlainSocketFactory.getSocketFactory(), httpPort)); schemeRegistry.register(new Scheme("https", sslSocketFactory, httpsPort)); return schemeRegistry; } /** * Creates a new AsyncHttpClient. * * @param schemeRegistry SchemeRegistry to be used */ public AsyncHttpClient(SchemeRegistry schemeRegistry) { // 最终调到的是这个版本。。。 BasicHttpParams httpParams = new BasicHttpParams(); // 接下来是设置各种参数。。。 ConnManagerParams.setTimeout(httpParams, connectTimeout); ConnManagerParams.setMaxConnectionsPerRoute(httpParams, new ConnPerRouteBean(maxConnections)); ConnManagerParams.setMaxTotalConnections(httpParams, DEFAULT_MAX_CONNECTIONS); HttpConnectionParams.setSoTimeout(httpParams, responseTimeout); HttpConnectionParams.setConnectionTimeout(httpParams, connectTimeout); HttpConnectionParams.setTcpNoDelay(httpParams, true); HttpConnectionParams.setSocketBufferSize(httpParams, DEFAULT_SOCKET_BUFFER_SIZE); HttpProtocolParams.setVersion(httpParams, HttpVersion.HTTP_1_1); ThreadSafeClientConnManager cm = new ThreadSafeClientConnManager(httpParams, schemeRegistry); // 初始化关键字段 threadPool = getDefaultThreadPool(); requestMap = Collections.synchronizedMap(new WeakHashMap<Context, List<RequestHandle>>()); clientHeaderMap = new HashMap<String, String>(); httpContext = new SyncBasicHttpContext(new BasicHttpContext()); httpClient = new DefaultHttpClient(cm, httpParams); httpClient.addRequestInterceptor(new HttpRequestInterceptor() { @Override public void process(HttpRequest request, HttpContext context) { if (!request.containsHeader(HEADER_ACCEPT_ENCODING)) { request.addHeader(HEADER_ACCEPT_ENCODING, ENCODING_GZIP); } for (String header : clientHeaderMap.keySet()) { if (request.containsHeader(header)) { Header overwritten = request.getFirstHeader(header); Log.d(LOG_TAG, String.format("Headers were overwritten! (%s | %s) overwrites (%s | %s)", header, clientHeaderMap.get(header), overwritten.getName(), overwritten.getValue()) ); //remove the overwritten header request.removeHeader(overwritten); } request.addHeader(header, clientHeaderMap.get(header)); } } }); httpClient.addResponseInterceptor(new HttpResponseInterceptor() { @Override public void process(HttpResponse response, HttpContext context) { final HttpEntity entity = response.getEntity(); if (entity == null) { return; } final Header encoding = entity.getContentEncoding(); if (encoding != null) { for (HeaderElement element : encoding.getElements()) { if (element.getName().equalsIgnoreCase(ENCODING_GZIP)) { response.setEntity(new InflatingEntity(entity)); break; } } } } }); httpClient.addRequestInterceptor(new HttpRequestInterceptor() { @Override public void process(final HttpRequest request, final HttpContext context) throws HttpException, IOException { AuthState authState = (AuthState) context.getAttribute(ClientContext.TARGET_AUTH_STATE); CredentialsProvider credsProvider = (CredentialsProvider) context.getAttribute( ClientContext.CREDS_PROVIDER); HttpHost targetHost = (HttpHost) context.getAttribute(ExecutionContext.HTTP_TARGET_HOST); if (authState.getAuthScheme() == null) { AuthScope authScope = new AuthScope(targetHost.getHostName(), targetHost.getPort()); Credentials creds = credsProvider.getCredentials(authScope); if (creds != null) { authState.setAuthScheme(new BasicScheme()); authState.setCredentials(creds); } } } }, 0); // 设置重试Handler，会在合适的情况下自动重试 httpClient.setHttpRequestRetryHandler(new RetryHandler(DEFAULT_MAX_RETRIES, DEFAULT_RETRY_SLEEP_TIME_MILLIS)); }
接下来重要的就是各种HTTP head、get、post、delete方法，它们最终调用的都是sendRequest方法，如下：
/** * Puts a new request in queue as a new thread in pool to be executed * * @param client HttpClient to be used for request, can differ in single requests * @param contentType MIME body type, for POST and PUT requests, may be null * @param context Context of Android application, to hold the reference of request * @param httpContext HttpContext in which the request will be executed * @param responseHandler ResponseHandler or its subclass to put the response into * @param uriRequest instance of HttpUriRequest, which means it must be of HttpDelete, * HttpPost, HttpGet, HttpPut, etc. * @return RequestHandle of future request process */ protected RequestHandle sendRequest(DefaultHttpClient client, HttpContext httpContext, HttpUriRequest uriRequest, String contentType, ResponseHandlerInterface responseHandler, Context context) { if (uriRequest == null) { throw new IllegalArgumentException("HttpUriRequest must not be null"); } if (responseHandler == null) { throw new IllegalArgumentException("ResponseHandler must not be null"); } if (responseHandler.getUseSynchronousMode()) { throw new IllegalArgumentException("Synchronous ResponseHandler used in AsyncHttpClient. You should create your response handler in a looper thread or use SyncHttpClient instead."); } if (contentType != null) { uriRequest.setHeader(HEADER_CONTENT_TYPE, contentType); } responseHandler.setRequestHeaders(uriRequest.getAllHeaders()); responseHandler.setRequestURI(uriRequest.getURI()); // 下面的这3行是重点，创建请求，提交请求到线程池，将请求包装到RequestHandle用于之后的取消、管理 AsyncHttpRequest request = newAsyncHttpRequest(client, httpContext, uriRequest, contentType, responseHandler, context); threadPool.submit(request); // 能submit说明request至少是个Runnable RequestHandle requestHandle = new RequestHandle(request); if (context != null) { // 如果Android context非空的话，做一些关联操作，后面可以通过context来取消request的执行 // Add request to request map List<RequestHandle> requestList = requestMap.get(context); synchronized (requestMap) { if (requestList == null) { requestList = Collections.synchronizedList(new LinkedList<RequestHandle>()); requestMap.put(context, requestList); } } if (responseHandler instanceof RangeFileAsyncHttpResponseHandler) ((RangeFileAsyncHttpResponseHandler) responseHandler).updateRequestHeaders(uriRequest); requestList.add(requestHandle); Iterator<RequestHandle> iterator = requestList.iterator(); while (iterator.hasNext()) { if (iterator.next().shouldBeGarbageCollected()) { iterator.remove(); // 清理已经完成/取消了的请求 } } } return requestHandle; }
看到了吧，发送请求的过程其实重点是创建请求，然后submit到线程池，剩下的事情就交给线程池自己处理了，我们只需要坐等被调用。
来看看创建请求的方法，代码如下：
/** * Instantiate a new asynchronous HTTP request for the passed parameters. * * @param client HttpClient to be used for request, can differ in single requests * @param contentType MIME body type, for POST and PUT requests, may be null * @param context Context of Android application, to hold the reference of request * @param httpContext HttpContext in which the request will be executed * @param responseHandler ResponseHandler or its subclass to put the response into * @param uriRequest instance of HttpUriRequest, which means it must be of HttpDelete, * HttpPost, HttpGet, HttpPut, etc. * @return AsyncHttpRequest ready to be dispatched */ protected AsyncHttpRequest newAsyncHttpRequest(DefaultHttpClient client, HttpContext httpContext, HttpUriRequest uriRequest, String contentType, ResponseHandlerInterface responseHandler, Context context) { return new AsyncHttpRequest(client, httpContext, uriRequest, responseHandler); }
紧接着我们看看AsyncHttpRequest的实现：
/** * Internal class, representing the HttpRequest, done in asynchronous manner */ public class AsyncHttpRequest implements Runnable { // 这就是submit到线程池的Runnable private final AbstractHttpClient client; private final HttpContext context; private final HttpUriRequest request; private final ResponseHandlerInterface responseHandler; private int executionCount; private boolean isCancelled; private boolean cancelIsNotified; private boolean isFinished; private boolean isRequestPreProcessed; public AsyncHttpRequest(AbstractHttpClient client, HttpContext context, HttpUriRequest request, ResponseHandlerInterface responseHandler) { this.client = client; this.context = context; this.request = request; this.responseHandler = responseHandler; } /** * This method is called once by the system when the request is about to be * processed by the system. The library makes sure that a single request * is pre-processed only once. * * Please note: pre-processing does NOT run on the main thread, and thus * any UI activities that you must perform should be properly dispatched to * the app‘s UI thread. * * @param request The request to pre-process */ public void onPreProcessRequest(AsyncHttpRequest request) { // default action is to do nothing... } /** * This method is called once by the system when the request has been fully * sent, handled and finished. The library makes sure that a single request * is post-processed only once. * * Please note: post-processing does NOT run on the main thread, and thus * any UI activities that you must perform should be properly dispatched to * the app‘s UI thread. * * @param request The request to post-process */ public void onPostProcessRequest(AsyncHttpRequest request) { // default action is to do nothing... } @Override public void run() { // 这是在线程池中执行的方法，我们重点看看 if (isCancelled()) { // 检测，如果已经取消了则直接返回，下面的代码有好多次做这个检测，因为你永远不知道什么时候会被取消 return; // 同时也说明了我们的Request是支持取消的 } // Carry out pre-processing for this request only once. if (!isRequestPreProcessed) { isRequestPreProcessed = true; onPreProcessRequest(this); // callback接口，在一次请求中只调用一次 } if (isCancelled()) { // 再次检查 return; } if (responseHandler != null) { responseHandler.sendStartMessage(); // 发送开始请求消息 } if (isCancelled()) { // 检查 return; } try { makeRequestWithRetries(); // 带自动retry机制的请求 } catch (IOException e) { if (!isCancelled() && responseHandler != null) { responseHandler.sendFailureMessage(0, null, null, e); // 在没取消的情况下，发送失败消息 } else { Log.e("AsyncHttpRequest", "makeRequestWithRetries returned error, but handler is null", e); } } if (isCancelled()) { // 检查again return; } if (responseHandler != null) { // 没取消的情况下，发送完成消息 responseHandler.sendFinishMessage(); } if (isCancelled()) { return; } // Carry out post-processing for this request. onPostProcessRequest(this); // 处理了请求之后的callback isFinished = true; // 设置为true表示这个请求执行完毕了 } private void makeRequest() throws IOException { // 发送一次请求 if (isCancelled()) { return; } // Fixes #115 if (request.getURI().getScheme() == null) { // subclass of IOException so processed in the caller throw new MalformedURLException("No valid URI scheme was provided"); } // 执行请求获得response HttpResponse response = client.execute(request, context); if (isCancelled() || responseHandler == null) { return; } // Carry out pre-processing for this response. responseHandler.onPreProcessResponse(responseHandler, response); // 处理response前 if (isCancelled()) { return; } // The response is ready, handle it. responseHandler.sendResponseMessage(response); // 发送获得的response if (isCancelled()) { return; } // Carry out post-processing for this response. responseHandler.onPostProcessResponse(responseHandler, response); // 处理response后 } private void makeRequestWithRetries() throws IOException { boolean retry = true; IOException cause = null; HttpRequestRetryHandler retryHandler = client.getHttpRequestRetryHandler(); try { while (retry) { // 注意这个循环，当retry为false的时候退出 try { makeRequest(); return; // 请求成功的话，直接返回 } catch (UnknownHostException e) { // switching between WI-FI and mobile data networks can cause a retry which then results in an UnknownHostException // while the WI-FI is initialising. The retry logic will be invoked here, if this is NOT the first retry // (to assist in genuine cases of unknown host) which seems better than outright failure cause = new IOException("UnknownHostException exception: " + e.getMessage()); retry = (executionCount > 0) && retryHandler.retryRequest(cause, ++executionCount, context); } catch (NullPointerException e) { // there‘s a bug in HttpClient 4.0.x that on some occasions causes // DefaultRequestExecutor to throw an NPE, see // http://code.google.com/p/android/issues/detail?id=5255 cause = new IOException("NPE in HttpClient: " + e.getMessage()); retry = retryHandler.retryRequest(cause, ++executionCount, context); } catch (IOException e) { if (isCancelled()) { // Eating exception, as the request was cancelled return; } cause = e; retry = retryHandler.retryRequest(cause, ++executionCount, context); }
// 各种异常的情况下，计算retry，看还是否需要retry if (retry && (responseHandler != null)) { // 需要retry的时候，发送retry消息并附带第几次retry了 responseHandler.sendRetryMessage(executionCount); } } } catch (Exception e) { // catch anything else to ensure failure message is propagated Log.e("AsyncHttpRequest", "Unhandled exception origin cause", e);
// 其他的所有不在上述catch里的异常都在这里统一包装成IOException，在最后抛出 cause = new IOException("Unhandled exception: " + e.getMessage()); } // cleaned up to throw IOException throw (cause); // 抛出，以便上层代码知道发生了什么 } public boolean isCancelled() { if (isCancelled) { sendCancelNotification(); } return isCancelled; } private synchronized void sendCancelNotification() { if (!isFinished && isCancelled && !cancelIsNotified) { cancelIsNotified = true; if (responseHandler != null) responseHandler.sendCancelMessage(); } } public boolean isDone() { return isCancelled() || isFinished; } public boolean cancel(boolean mayInterruptIfRunning) { isCancelled = true; request.abort(); return isCancelled(); } }
紧接着，我们大概提下RequestHandle，它只是一个持有AsyncHttpRequest对象的弱引用，其方法内部都delegate给了AsyncHttpRequest，
非常简单，感兴趣的同学可自行阅读。
看完了Request，接下来该看看各种Response了，他们都实现了ResponseHandlerInterface接口，这里我们重点看下AsyncHttpResponseHandler，
因为它是后面所有更具体的子类的基础，其ctor代码如下：
/** * Creates a new AsyncHttpResponseHandler */ public AsyncHttpResponseHandler() { // 不指定looper this(null); } /** * Creates a new AsyncHttpResponseHandler with a user-supplied looper. If * the passed looper is null, the looper attached to the current thread will * be used. * * @param looper The looper to work with */ public AsyncHttpResponseHandler(Looper looper) { // 如果没指定looper的话，会用当前线程的looper顶替 this.looper = looper == null ? Looper.myLooper() : looper; // Use asynchronous mode by default. setUseSynchronousMode(false); // 默认是异步的方式，这里异步的意思是指对response的处理发生在与looper } // 关联的线程中，而不是请求发生的线程池里的线程中 @Override public void setUseSynchronousMode(boolean sync) { // A looper must be prepared before setting asynchronous mode. if (!sync && this.looper == null) { sync = true; // 一种错误的情况，强制使用同步mode Log.w(LOG_TAG, "Current thread has not called Looper.prepare(). Forcing synchronous mode."); } // If using asynchronous mode. if (!sync && handler == null) { // 初始化handler // Create a handler on current thread to submit tasks handler = new ResponderHandler(this, this.looper); } else if (sync && handler != null) { // TODO: Consider adding a flag to remove all queued messages. handler = null; } useSynchronousMode = sync; }
一般来说，我们会直接在UI线程中调用无参版本的ctor，也就是说response是和UI线程关联的，所有对其的处理handleMessage是发生
在UI线程中的。如果你想用response的结果来更新UI则这是正确的方式。
接着我们看看和处理response相关的代码：
/** * Avoid leaks by using a non-anonymous handler class. */ private static class ResponderHandler extends Handler { private final AsyncHttpResponseHandler mResponder; ResponderHandler(AsyncHttpResponseHandler mResponder, Looper looper) { super(looper); this.mResponder = mResponder; } @Override public void handleMessage(Message msg) { // 一个简单的Handler，其handleMessage delegate给了mResponder mResponder.handleMessage(msg); } } // Methods which emulate android‘s Handler and Message methods protected void handleMessage(Message message) { // 对各种message的处理，回调各种onXXX方法 Object[] response; switch (message.what) { case SUCCESS_MESSAGE: response = (Object[]) message.obj; if (response != null && response.length >= 3) { onSuccess((Integer) response[0], (Header[]) response[1], (byte[]) response[2]); } else { Log.e(LOG_TAG, "SUCCESS_MESSAGE didn‘t got enough params"); } break; case FAILURE_MESSAGE: response = (Object[]) message.obj; if (response != null && response.length >= 4) { onFailure((Integer) response[0], (Header[]) response[1], (byte[]) response[2], (Throwable) response[3]); } else { Log.e(LOG_TAG, "FAILURE_MESSAGE didn‘t got enough params"); } break; case START_MESSAGE: onStart(); break; case FINISH_MESSAGE: onFinish(); break; case PROGRESS_MESSAGE: response = (Object[]) message.obj; if (response != null && response.length >= 2) { try { onProgress((Integer) response[0], (Integer) response[1]); } catch (Throwable t) { Log.e(LOG_TAG, "custom onProgress contains an error", t); } } else { Log.e(LOG_TAG, "PROGRESS_MESSAGE didn‘t got enough params"); } break; case RETRY_MESSAGE: response = (Object[]) message.obj; if (response != null && response.length == 1) { onRetry((Integer) response[0]); } else { Log.e(LOG_TAG, "RETRY_MESSAGE didn‘t get enough params"); } break; case CANCEL_MESSAGE: onCancel(); break; } } protected void sendMessage(Message msg) { if (getUseSynchronousMode() || handler == null) { handleMessage(msg); // 如果是同步的方式，则handleMessage发生在调用sendMessage的线程中 } else if (!Thread.currentThread().isInterrupted()) { // do not send messages if request has been cancelled handler.sendMessage(msg); // 否则发生在与handler关联的线程中，一般多为UI线程 } }
代码中各种sendXXXMessage都会调用这里的sendMessage方法，只是构造的msg的what、obj不同而已。而sendXXXMessage方法
会在request的不同阶段自动被调用，详见AsyncHttpRequest中。下一步我们看眼对response的解析过程，代码如下：
@Override public void sendResponseMessage(HttpResponse response) throws IOException { // do not process if request has been cancelled if (!Thread.currentThread().isInterrupted()) { StatusLine status = response.getStatusLine(); byte[] responseBody; responseBody = getResponseData(response.getEntity()); // 将response解析成字节数组 // additional cancellation check as getResponseData() can take non-zero time to process if (!Thread.currentThread().isInterrupted()) { if (status.getStatusCode() >= 300) { // 标志失败的情况 sendFailureMessage(status.getStatusCode(), response.getAllHeaders(), responseBody, new HttpResponseException(status.getStatusCode(), status.getReasonPhrase())); } else { // 成功的情况 sendSuccessMessage(status.getStatusCode(), response.getAllHeaders(), responseBody); } } } } /** * Returns byte array of response HttpEntity contents * * @param entity can be null * @return response entity body or null * @throws java.io.IOException if reading entity or creating byte array failed */ byte[] getResponseData(HttpEntity entity) throws IOException { byte[] responseBody = null; if (entity != null) { InputStream instream = entity.getContent(); // 从entity中读取字节流 if (instream != null) { long contentLength = entity.getContentLength(); if (contentLength > Integer.MAX_VALUE) { throw new IllegalArgumentException("HTTP entity too large to be buffered in memory"); } int buffersize = (contentLength <= 0) ? BUFFER_SIZE : (int) contentLength; try { ByteArrayBuffer buffer = new ByteArrayBuffer(buffersize); try { byte[] tmp = new byte[BUFFER_SIZE]; int l, count = 0; // do not send messages if request has been cancelled while ((l = instream.read(tmp)) != -1 && !Thread.currentThread().isInterrupted()) { count += l; buffer.append(tmp, 0, l); sendProgressMessage(count, (int) (contentLength <= 0 ? 1 : contentLength)); } } finally { AsyncHttpClient.silentCloseInputStream(instream); AsyncHttpClient.endEntityViaReflection(entity); } responseBody = buffer.toByteArray(); } catch (OutOfMemoryError e) { System.gc(); throw new IOException("File too large to fit into available memory"); } } } return responseBody; }
onXXX方法除了onSuccess和onFailure外都做了默认实现即啥也不做，所以继承至它的子类至少要实现这2个方法，其他的方法你可以选择性实现。
接下来我们看看TextHttpResponseHandler子类的实现，关键代码如下：
@Override // 对上述2个方法的重载，其中将byte[]通过getResponseString方法转化成了String对象 public void onSuccess(int statusCode, Header[] headers, byte[] responseBytes) { onSuccess(statusCode, headers, getResponseString(responseBytes, getCharset())); } @Override public void onFailure(int statusCode, Header[] headers, byte[] responseBytes, Throwable throwable) { onFailure(statusCode, headers, getResponseString(responseBytes, getCharset()), throwable); } /** * Attempts to encode response bytes as string of set encoding * * @param charset charset to create string with * @param stringBytes response bytes * @return String of set encoding or null */ public static String getResponseString(byte[] stringBytes, String charset) { try { String toReturn = (stringBytes == null) ? null : new String(stringBytes, charset); if (toReturn != null && toReturn.startsWith(UTF8_BOM)) { return toReturn.substring(1); } return toReturn; } catch (UnsupportedEncodingException e) { Log.e(LOG_TAG, "Encoding response into string failed", e); return null; } }
说白了，也就是在父类基础上多了一层处理，将byte[]根据特定的编码转化成String而已，类似的JsonHttpResponseHandler又在此基础上
将String转化成JSONObject或JSONArray，细节不赘述。
ResponseHandler介绍完了，这里我们提下RetryHandler，这个类也很简单，根据内部的白/黑名单等规则来确定是否要retry。
AsyncHttpClient当然也提供了对Cookie的支持，默认是保存在Android的SharedPreferences中，具体代码见PersistentCookieStore。
还有一个功能丰富的RequestParams类，据此你不仅可以为GET/POST方法提供参数，甚至你可以上传本地文件到server端。
到此为止，AsyncHttpClient关键部分的代码已基本分析完毕了，剩下的还需要大家自己在项目中多多实践，enjoy。。。
AsyncHttpClient 源码分析,布布扣,bubuko.com
原文：http://www.cnblogs.com/xiaoweiz/p/3918042.html