背景
android开发过程中网络请求作为最重要的组成部分之一,然而对于大部分android开发者在网络请求上有太多疑惑,不知道如何去选型?通过原生的HttpClient、HttpUrlConnection封装?还是通过第三方框架再封装?笔者以为采用广泛被使用的第三方网络框架再封装为上策,因为这些网络框架如retrofit、okhttp、volley等是被全球android开发者维护着,无论在功能上、性能上、还是代码简洁性都相对于自己通过原生实现的给力.
目的
致力封装一个简洁、实用、易移植的网络框架模块.
正题
今天笔者就给大家基于retrofit + okhttp + gson 封装一个通用易懂的网络框架模块.
首先我们需要在gradle文件中将第三方依赖引入项目,代码如下:
dependencies { // retrofit + okhttp + gson compile 'com.squareup.retrofit2:retrofit:2.1.0' compile 'com.squareup.okhttp3:okhttp:3.4.1' compile 'com.google.code.gson:gson:2.7' compile 'com.squareup.retrofit2:converter-gson:2.1.0'}
接着开始我们的封装之路......
首先我们需要写一个参数常量类,用于定义一些常量,如请求Url地址、接口返回信息,代码如下:
/** * @className: InterfaceParameters * @classDescription: 参数配置 * @author: leibing * @createTime: 2016/8/30 */public class InterfaceParameters { // 请求URL public final static String REQUEST_HTTP_URL = BuildConfig.API_URL; // 接口返回结果名称 public final static String INFO = "info"; // 接口返回错误码 public final static String ERROR_CODE = "errorcode"; // 接口返回错误信息 public final static String ERROR_MSG = "errormsg"; }
然后写一个网络请求封装类JkApiRequest.class,采用单例的方式,配置网络请求参数以及返回网络请求api实例,代码如下:
/** * @className:JkApiRequest * @classDescription:网络请求 * @author: leibing * @createTime: 2016/8/30 */public class JkApiRequest { // sington private static JkApiRequest instance; // Retrofit object private Retrofit retrofit; /** * Constructor * @author leibing * @createTime 2016/08/30 * @lastModify 2016/08/30 * @param * @return */ private JkApiRequest(){ OkHttpClient client = new OkHttpClient.Builder() .addInterceptor(new OkHttpInterceptor()) .build(); retrofit = new Retrofit.Builder() .baseUrl(InterfaceParameters.REQUEST_HTTP_URL) .addConverterFactory(JkApiConvertFactory.create()) .client(client) .build(); } /** * sington * @author leibing * @createTime 2016/08/30 * @lastModify 2016/08/30 * @param * @return */ public static JkApiRequest getInstance(){ if (instance == null){ instance = new JkApiRequest(); } return instance; } /** * create api instance * @author leibing * @createTime 2016/08/30 * @lastModify 2016/08/30 * @param service api class * @return */ public <T> T create(Class<T> service) { return retrofit.create(service); } }
上面代码有两个网络请求参数需要注意, OkHttpInterceptor 、JkApiConvertFactory.
OkHttpInterceptor 作为网络请求拦截器,可以拦截请求的数据以及响应的数据,有助于我们排查问题,而JkApiConvertFactory 是作为Convert工厂,这里我所做的就是解析返回来的数据,将数据进行Gson处理就是在这里面.
OkHttpInterceptor代码如下:
/** * @className: OkHttpInterceptor * @classDescription: Http拦截器 * @author: leibing * @createTime: 2016/08/30 */public class OkHttpInterceptor implements Interceptor { private static final Charset UTF8 = Charset.forName("UTF-8"); @Override public Response intercept(Chain chain) throws IOException { Request request = chain.request(); // 获得Connection,内部有route、socket、handshake、protocol方法 Connection connection = chain.connection(); // 如果Connection为null,返回HTTP_1_1,否则返回connection.protocol() Protocol protocol = connection != null ? connection.protocol() : Protocol.HTTP_1_1; // 比如: --> POST http://121.40.227.8:8088/api http/1.1 String requestStartMessage = "--> " + request.method() + ' ' + request.url() + ' ' + protocol; System.out.println("ddddddddddddddddddd requestStartMessage = " + requestStartMessage); // 打印 Response Response response; try { response = chain.proceed(request); } catch (Exception e) { throw e; } ResponseBody responseBody = response.body(); long contentLength = responseBody.contentLength(); if (bodyEncoded(response.headers())) { } else { BufferedSource source = responseBody.source(); source.request(Long.MAX_VALUE); // Buffer the entire body. Buffer buffer = source.buffer(); Charset charset = UTF8; if (contentLength != 0) { // 获取Response的body的字符串 并打印 System.out.println("ddddddddddddddddddddddddd response = " + buffer.clone().readString(charset)); } } return response; } private boolean bodyEncoded(Headers headers) { String contentEncoding = headers.get("Content-Encoding"); return contentEncoding != null && !contentEncoding.equalsIgnoreCase("identity"); } }
JkApiConvertFactory代码如下:
/** * @className: JkApiConvertFactory * @classDescription: this converter decode the response. * @author: leibing * @createTime: 2016/8/30 */public class JkApiConvertFactory extends Converter.Factory{ public static JkApiConvertFactory create() { return create(new Gson()); } public static JkApiConvertFactory create(Gson gson) { return new JkApiConvertFactory(gson); } private JkApiConvertFactory(Gson gson) { if (gson == null) throw new NullPointerException("gson == null"); } @Override public Converter<ResponseBody, ?> responseBodyConverter(Type type, Annotation[] annotations, Retrofit retrofit) { return new GsonResponseBodyConverter<>(type); } final class GsonResponseBodyConverter<T> implements Converter<ResponseBody, T> { private final Type type; GsonResponseBodyConverter(Type type) { this.type = type; } @Override public T convert(ResponseBody value) throws IOException { BaseResponse baseResponse; String reString; try { reString = value.string(); // 解析Json数据返回TransData对象 TransData transData = TransUtil.getResponse(reString); try { if (transData.getErrorcode().equals("0") && !TextUtils.isEmpty(transData.getInfo())) { baseResponse = new Gson().fromJson(transData.getInfo(), type); baseResponse.setSuccess(transData.getErrorcode().equals("0")); baseResponse.setInfo(transData.getInfo()); baseResponse.setInfo(transData.getInfo()); } else { baseResponse = (BaseResponse) StringUtil.getObject(((Class) type).getName()); baseResponse.setSuccess(transData.getErrorcode().equals("0")); baseResponse.setInfo(transData.getInfo()); baseResponse.setInfo(transData.getInfo()); } return (T)baseResponse; } catch (Exception e) { e.printStackTrace(); } } catch (Exception e) { e.printStackTrace(); } //从不返回一个空的Response. baseResponse = (BaseResponse) StringUtil.getObject(((Class) type).getName()); try { baseResponse.setSuccess(false); //JkApiConvertFactory can not recognize the response! baseResponse.setErrormsg(""); } catch (Exception e) { e.printStackTrace(); } finally { return (T)baseResponse; } } } }
接着我们写api接口,我这里是将对应的接口封装到对应的类中,这样当前api类中接口名称可以统一起来,请求api的方法也写入当前api类,这样做既能解耦又能减少在使用处的冗余代码,代码如下:
/** * @className: ApiLogin * @classDescription: 登陆api * @author: leibing * @createTime: 2016/8/12 */public class ApiLogin { // api private ApiStore mApiStore; /** * Constructor * @author leibing * @createTime 2016/08/30 * @lastModify 2016/08/30 * @param * @return */ public ApiLogin() { // 初始化api mApiStore = JkApiRequest.getInstance().create(ApiStore.class); } /** * 登录 * @author leibing * @createTime 2016/08/30 * @lastModify 2016/08/30 * @param username 用户名 * @param password 密码 * @param callback 回调 * @return */ public void Login(String username, String password, ApiCallback<LoginResponse> callback){ Call<LoginResponse> mCall = mApiStore.login(URLEncoder.encode(username), password); mCall.enqueue(new JkApiCallback<LoginResponse>(callback)); } /** * @interfaceName: ApiStore * @interfaceDescription: 登录模块api接口 * @author: leibing * @createTime: 2016/08/30 */ private interface ApiStore { @GET("app/User/Login") Call<LoginResponse> login( @Query("username") String username, @Query("userpass") String userpass); } }
从上面代码我们看到ApiCallback和JkApiCallback两个回调接口,这两者区别在哪呢?观察仔细的童鞋会发现这个问题.ApiCallback接口是作为通用接口,而JkApiCallback是作为一个接口封装类,针对不同数据情景,做不同回调.
ApiCallback代码如下:
/** * @className: ApiCallback * @classDescription: Api回调 * @author: leibing * @createTime: 2016/08/30 */public interface ApiCallback<T> { // 请求数据成功 void onSuccess(T response); // 请求数据错误 void onError(String err_msg); // 网络请求失败 void onFailure(); }
JkApiCallback代码如下:
public class JkApiCallback<T> implements Callback <T>{ // 回调 private ApiCallback<T> mCallback; /** * Constructor * @author leibing * @createTime 2016/08/30 * @lastModify 2016/08/30 * @param mCallback 回调 * @return */ public JkApiCallback(ApiCallback<T> mCallback){ this.mCallback = mCallback; } @Override public void onResponse(Call<T> call, Response<T> response) { if (mCallback == null){ throw new NullPointerException("mCallback == null"); } if (response != null && response.body() != null){ if (((BaseResponse)response.body()).isSuccess()){ mCallback.onSuccess((T)response.body()); }else { mCallback.onError(((BaseResponse) response.body()).getErrormsg()); } }else { mCallback.onFailure(); } } @Override public void onFailure(Call<T> call, Throwable t) { if (mCallback == null){ throw new NullPointerException("mCallback == null"); } mCallback.onFailure(); } }
接下来我们写Response类,用于接收Gson解析回来的数据,这个只需写json数据信息里面的字段.代码如下:
/** * @className: LoginResponse * @classDescription: 获取登录返回的信息 * @author: leibing * @createTime: 2016/08/30 */public class LoginResponse extends BaseResponse implements Serializable{ // 序列化UID 用于反序列化 private static final long serialVersionUID = 4863726647304575308L; // token public String accesstoken; }
阅读仔细的童鞋会发现,在Convert工厂中Gson解析时,用到了一个BaseResponse,这个响应类是作为基类,就是服务端返回的固定json数据格式,因为每个项目返回的固定格式可能不一样,所以只需改这里,就能定制对应项目的网络框架.
BaseResponse代码如下:
/** * @className: BaseResponse * @classDescription: 网络请求返回对象公共抽象类 * @author: leibing * @createTime: 2016/08/30 */public class BaseResponse implements Serializable { // 序列化UID 用于反序列化 private static final long serialVersionUID = 234513596098152098L; // 是否成功 private boolean isSuccess; // 数据 public String info; // 错误消息 public String errormsg; public boolean isSuccess() { return isSuccess; } public void setSuccess(boolean success) { isSuccess = success; } public String getInfo() { return info; } public void setInfo(String info) { this.info = info; } public String getErrormsg() { return errormsg; } public void setErrormsg(String errormsg) { this.errormsg = errormsg; } }
一个简洁、实用、易移植的网络框架模块封装完毕.