Android开源项目xUtils HttpUtils模块分析

xUtils是github上的一个Android开源工具项目，其中HttpUtils模块是处理网络连接部分，刚好最近想整理下Android网络编程知识，今天学习下xUtils中HttpUtils.

xUtils项目地址：https://github.com/wyouflf/xUtils

先看看分析的时序图，了解下基本的过程

1. Activity创建HttpUtils对象

HttpUtils http = new HttpUtils()；

查看HttpUtils类的构造函数

 public HttpUtils() {
        this(HttpUtils.DEFAULT_CONN_TIMEOUT);
    }

    public HttpUtils(int connTimeout) {
        HttpParams params = new BasicHttpParams();

        ConnManagerParams.setTimeout(params, connTimeout);
        HttpConnectionParams.setSoTimeout(params, connTimeout);
        HttpConnectionParams.setConnectionTimeout(params, connTimeout);

        ConnManagerParams.setMaxConnectionsPerRoute(params, new ConnPerRouteBean(10));
        ConnManagerParams.setMaxTotalConnections(params, 10);

        HttpConnectionParams.setTcpNoDelay(params, true);
        HttpConnectionParams.setSocketBufferSize(params, 1024 * 8);
        HttpProtocolParams.setVersion(params, HttpVersion.HTTP_1_1);

        SchemeRegistry schemeRegistry = new SchemeRegistry();
        schemeRegistry.register(new Scheme("http", PlainSocketFactory.getSocketFactory(), 80));
        schemeRegistry.register(new Scheme("https", DefaultSSLSocketFactory.getSocketFactory(), 443));

        httpClient = new DefaultHttpClient(new ThreadSafeClientConnManager(params, schemeRegistry), params);

        httpClient.setHttpRequestRetryHandler(new RetryHandler(DEFAULT_RETRY_TIMES));

        httpClient.addRequestInterceptor(new HttpRequestInterceptor() {
            @Override
            public void process(org.apache.http.HttpRequest httpRequest, HttpContext httpContext) throws org.apache.http.HttpException, IOException {
                if (!httpRequest.containsHeader(HEADER_ACCEPT_ENCODING)) {
                    httpRequest.addHeader(HEADER_ACCEPT_ENCODING, ENCODING_GZIP);
                }
            }
        });

        httpClient.addResponseInterceptor(new HttpResponseInterceptor() {
            @Override
            public void process(HttpResponse response, HttpContext httpContext) throws org.apache.http.HttpException, IOException {
                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("gzip")) {
                            response.setEntity(new GZipDecompressingEntity(response.getEntity()));
                            return;
                        }
                    }
                }
            }
        });
    }

这里主要是设置HttpParams参数，然后创建httpClient对象。

注意这个类ThreadSafeClientConnManager，它主要是为了使用线程安全的连接管理来创建HttpClient。

不过这里就有个疑问了，之前看资料了解到一般创建HttpClient都是用的 单例模式，说是一个httpClient就相当于是一个小型的浏览器，如果创建多个httpClient就很消耗资源了，我看了这个开源项目给的demo，是创建一个请求就创建一个HttpClient, 到时跟作者联系看看是什么回事。

2.发送请求

 http.send(HttpRequest.HttpMethod.GET,
                "http://www.baidu.com",
                new RequestCallBack() {


                    @Override
                    public void onStart() {
                        resultText.setText("conn...");
                    }


                    @Override
                    public void onLoading(long total, long current, boolean isUploading) {
                        resultText.setText(current + "/" + total);
                    }


                    @Override
                    public void onSuccess(ResponseInfo responseInfo) {
                        resultText.setText("response:" + responseInfo.result);
                    }




                    @Override
                    public void onFailure(HttpException error, String msg) {
                        resultText.setText(msg);
                    }
                });

调用send方法发生请求，

HttpRequest.HttpMethod.GET指明请求的方式，

"http://www.baidu.com"请求的地址，

new RequestCallBack()请求的回调函数，这里面四个方法方便开发者处理请求的各个阶段的结果。

3. http.send()

    public  HttpHandler send(HttpRequest.HttpMethod method, String url,
                                   RequestCallBack callBack) {
        return send(method, url, null, callBack);
    }

    public  HttpHandler send(HttpRequest.HttpMethod method, String url, RequestParams params,
                                   RequestCallBack callBack) {
        if (url == null) throw new IllegalArgumentException("url may not be null");

        HttpRequest request = new HttpRequest(method, url);
        return sendRequest(request, params, callBack);
    }
    private  HttpHandler sendRequest(HttpRequest request, RequestParams params, RequestCallBack callBack) {

        HttpHandler handler = new HttpHandler(httpClient, httpContext, responseTextCharset, callBack);

        handler.setExpiry(currentRequestExpiry);
        handler.setHttpRedirectHandler(httpRedirectHandler);
        request.setRequestParams(params, handler);

        handler.executeOnExecutor(executor, request);
        return handler;
    }

查看httpUtils的send函数，发现最后会调用sendRequest函数

在sendRequest里创建HttpHandler对象

4. HttpHandler

HttpHandler extends CompatibleAsyncTask implements RequestCallBackHandler

参看httpHandler发现它继承CompatibleAsyncTask

5.CompatibleAsyncTask

查看CompatibleAsyncTask ，发现它是A compatible AsyncTask for android2.2.你懂得

6.handler.executeOnExecutor(executor, request)

在第3步里创建完httpHandler后，调用handler.executeOnExecutor(executor, request)，而通过第4步了解到httpHandler继承CompatiableAsyncTask, 就先去看看doInBackground里做了什么事情。

7.doInBackground(Object... params)

            //先处理传递进来的params
            
            this.publishProgress(UPDATE_START);


            lastUpdateTime = SystemClock.uptimeMillis();


            ResponseInfo responseInfo = sendRequest(request);
            if (responseInfo != null) {
                this.publishProgress(UPDATE_SUCCESS, responseInfo);
                return null;
            }

先处理传递进来的params，调用publishProgress更新下当前的状态，然后调用sendRequest

8.sendRequest(HttpRequestBase request)

  private ResponseInfo sendRequest(HttpRequestBase request) throws HttpException {


        HttpRequestRetryHandler retryHandler = client.getHttpRequestRetryHandler();
        while (true) {

                requestMethod = request.getMethod();
                if (HttpUtils.sHttpCache.isEnabled(requestMethod)) {
                    String result = HttpUtils.sHttpCache.get(requestUrl);
                    if (result != null) {
                        return new ResponseInfo(null, (T) result, true);
                    }
                }


                ResponseInfo responseInfo = null;
                if (!isCancelled()) {
                    HttpResponse response = client.execute(request, context);
                    responseInfo = handleResponse(response);
                }
                return responseInfo;
            } catch (Exception e) {
                exception = e;
                retry = retryHandler.retryRequest(exception, ++retriedCount, context);
            } 
            if (!retry) {
                throw new HttpException(exception);
            }
        }
    }

这个方法仔细看看，

先获取下client.getHttpRequestRetryHandler()，获取retry的设置

requestMethod = request.getMethod();
                if (HttpUtils.sHttpCache.isEnabled(requestMethod)) {
                    String result = HttpUtils.sHttpCache.get(requestUrl);
                    if (result != null) {
                        return new ResponseInfo(null, (T) result, true);
                    }
                }

如果使用了缓存则通过requestUrl去httpCache去获取，获取到了则创建ResponseInfo对象

如果没有缓存

HttpResponse response = client.execute(request, context);
                    responseInfo = handleResponse(response);

调用httpClient执行http请求，获取到得结果交由handleResponse处理

如果之前的处理出现异常则

 retry = retryHandler.retryRequest(exception, ++retriedCount, context);

调用retry机制，直到有结果，或者超过retry的次数

9.handleResponse()

第8步的时候，如果client执行获取到结果则调用handleResponse(HttpResponse response)处理结果

 ResponseInfo handleResponse(HttpResponse response) throws HttpException, IOException {

        StatusLine status = response.getStatusLine();
        int statusCode = status.getStatusCode();
        if (statusCode < 300) {
  
                    result = mStringDownloadHandler.handleEntity(entity, this, charset);
                    if (HttpUtils.sHttpCache.isEnabled(requestMethod)) {
                        HttpUtils.sHttpCache.put(requestUrl, (String) result, expiry);
                    }
                
            }
            return new ResponseInfo(response, (T) result, false);
        } else if (statusCode == 301 || statusCode == 302) {
            if (httpRedirectHandler == null) {
                httpRedirectHandler = new DefaultHttpRedirectHandler();
            }
            HttpRequestBase request = httpRedirectHandler.getDirectRequest(response);
            if (request != null) {
                return this.sendRequest(request);
            }
        } 
        return null;
    }

这个方法主要根据返回的statuscode处理，<300将结果存在HttpCache里，301或者302则处理重定向

10.publishProgress(UPDATE_SUCCESS, responseInfo)

在获得ResponseInfo后，调用 publishProgress(UPDATE_SUCCESS, responseInfo)方法，最后会调用onProgressUpdate方法

protected void onProgressUpdate(Object... values) {
            case UPDATE_SUCCESS:
                if (values.length != 2) return;
                this.state = State.SUCCESS;
                callback.onSuccess((ResponseInfo) values[1]);
                break;
            default:
                break;
        }
    }

这onProgressUpdate里发现最终调用第2步传进来的callback

整个的调用过程基本上是这样。

1.创建httputils时创建httpClient,调用send发送请求

2. 调用send时，创建httpHandler，此类继承CompatibleAsyncTask

3.在httpHandler的doInBackground真正的处理http请求，此时会判断是否有缓存，获取结果后，通过回调处理结果