一些Spring MVC的使用技巧
_Arvin
8年前
<h3>APP服务端的Token验证</h3> <p>通过拦截器对使用了 @Authorization 注解的方法进行请求拦截,从http header中取出token信息,验证其是否合法。非法直接返回401错误,合法将token对应的user key存入request中后继续执行。具体实现代码:</p> <pre> <code class="language-java">public boolean preHandle(HttpServletRequest request, HttpServletResponse response, Object handler) throws Exception { //如果不是映射到方法直接通过 if (!(handler instanceof HandlerMethod)) { return true; } HandlerMethod handlerMethod = (HandlerMethod) handler; Method method = handlerMethod.getMethod(); //从header中得到token String token = request.getHeader(httpHeaderName); if (token != null && token.startsWith(httpHeaderPrefix) && token.length() > 0) { token = token.substring(httpHeaderPrefix.length()); //验证token String key = manager.getKey(token); if (key != null) { //如果token验证成功,将token对应的用户id存在request中,便于之后注入 request.setAttribute(REQUEST_CURRENT_KEY, key); return true; } } //如果验证token失败,并且方法注明了Authorization,返回401错误 if (method.getAnnotation(Authorization.class) != null) { response.setStatus(HttpServletResponse.SC_UNAUTHORIZED); response.setCharacterEncoding("gbk"); response.getWriter().write(unauthorizedErrorMessage); response.getWriter().close(); return false; } //为了防止以某种直接在REQUEST_CURRENT_KEY写入key,将其设为null request.setAttribute(REQUEST_CURRENT_KEY, null); return true; }</code></pre> <p>通过拦截器后,使用解析器对修饰了 @CurrentUser 的参数进行注入。从request中取出之前存入的user key,得到对应的user对象并注入到参数中。具体实现代码:</p> <pre> <code class="language-java">@Override public boolean supportsParameter(MethodParameter parameter) { Class clazz; try { clazz = Class.forName(userModelClass); } catch (ClassNotFoundException e) { return false; } //如果参数类型是User并且有CurrentUser注解则支持 if (parameter.getParameterType().isAssignableFrom(clazz) && parameter.hasParameterAnnotation(CurrentUser.class)) { return true; } return false; } @Override public Object resolveArgument(MethodParameter parameter, ModelAndViewContainer mavContainer, NativeWebRequest webRequest, WebDataBinderFactory binderFactory) throws Exception { //取出鉴权时存入的登录用户Id Object object = webRequest.getAttribute(AuthorizationInterceptor.REQUEST_CURRENT_KEY, RequestAttributes.SCOPE_REQUEST); if (object != null) { String key = String.valueOf(object); //从数据库中查询并返回 Object userModel = userModelRepository.getCurrentUser(key); if (userModel != null) { return userModel; } //有key但是得不到用户,抛出异常 throw new MissingServletRequestPartException(AuthorizationInterceptor.REQUEST_CURRENT_KEY); } //没有key就直接返回null return null; }</code></pre> <p>详细分析: <a href="/misc/goto?guid=4959674954579023504" rel="nofollow,noindex">RESTful登录设计(基于Spring及Redis的Token鉴权)</a></p> <p>源码见: <a href="/misc/goto?guid=4959674954662489548" rel="nofollow,noindex">ScienJus/spring-restful-authorization</a></p> <p>封装好的工具类: <a href="/misc/goto?guid=4959674954753452102" rel="nofollow,noindex">ScienJus/spring-authorization-manager</a></p> <h3>使用别名接受对象的参数</h3> <p>请求中的参数名和代码中定义的参数名不同是很常见的情况,对于这种情况Spring提供了几种原生的方法:</p> <p>对于 @RequestParam 可以直接指定value值为别名( @RequestHeader 也是一样),例如:</p> <pre> <code class="language-java">public String home(@RequestParam("user_id") long userId) { return "hello " + userId; }</code></pre> <p>对于 @RequestBody ,由于其使使用Jackson将Json转换为对象,所以可以使用 @JsonProperty 的value指定别名,例如:</p> <pre> <code class="language-java">public String home(@RequestBody User user) { return "hello " + user.getUserId(); } class User { @JsonProperty("user_id") private long userId; }</code></pre> <p>但是使用对象的属性接受参数时,就无法直接通过上面的办法指定别名了,例如:</p> <pre> <code class="language-java">public String home(User user) { return "hello " + user.getUserId(); }</code></pre> <p>这时候需要使用DataBinder手动绑定属性和别名,我在StackOverFlow上找到的 <a href="/misc/goto?guid=4959674954836689754" rel="nofollow,noindex">这篇文章</a> 是个不错的办法,这里就不重复造轮子了。</p> <h3>关闭默认通过请求的后缀名判断Content-Type</h3> <p>之前接手的项目的开发习惯是使用.html作为请求的后缀名,这在Struts2上是没有问题的(因为本身Struts2处理Json的几种方法就都很烂)。但是我接手换成Spring MVC后,使用 @ResponseBody 返回对象时就会报找不到转换器错误。</p> <p>这是因为Spring MVC默认会将后缀名为.html的请求的Content-Type认为是 text/html ,而 @ResponseBody 返回的Content-Type是 application/json ,没有任何一种转换器支持这样的转换。所以需要手动将通过后缀名判断Content-Type的设置关掉,并将默认的Content-Type设置为 application/json :</p> <pre> <code class="language-java">@Configuration public class WebMvcConfig extends WebMvcConfigurerAdapter { @Override public void configureContentNegotiation(ContentNegotiationConfigurer configurer) { configurer.favorPathExtension(false). defaultContentType(MediaType.APPLICATION_JSON); } }</code></pre> <h3>更改默认的Json序列化方案</h3> <p>项目中有时候会有自己独特的Json序列化方案,例如比较常用的使用 0 / 1 替代 false / true ,或是通过 "" 代替 null ,由于 @ResponseBody 默认使用的是 MappingJackson2HttpMessageConverter ,只需要将自己实现的 ObjectMapper 传入这个转换器:</p> <pre> <code class="language-java">public class CustomObjectMapper extends ObjectMapper { public CustomObjectMapper() { super(); this.getSerializerProvider().setNullValueSerializer(new JsonSerializer<Object>() { @Override public void serialize(Object value, JsonGenerator jgen, SerializerProvider provider) throws IOException { jgen.writeString(""); } }); SimpleModule module = new SimpleModule(); module.addSerializer(boolean.class, new JsonSerializer<Boolean>() { @Override public void serialize(Boolean value, JsonGenerator jgen, SerializerProvider provider) throws IOException { jgen.writeNumber(value ? 1 : 0); } }); this.registerModule(module); } }</code></pre> <h3>自动加密/解密请求中的Json</h3> <p>涉及到 @RequestBody 和 @ResponseBody 的类型转换问题一般都在 MappingJackson2HttpMessageConverter 中解决,想要自动加密/解密只需要继承这个类并重写 readInternal / writeInternal 方法即可:</p> <pre> <code class="language-java">@Override protected Object readInternal(Class<?> clazz, HttpInputMessage inputMessage) throws IOException, HttpMessageNotReadableException { //解密 String json = AESUtil.decrypt(inputMessage.getBody()); JavaType javaType = getJavaType(clazz, null); //转换 return this.objectMapper.readValue(json, javaType); } @Override protected void writeInternal(Object object, HttpOutputMessage outputMessage) throws IOException, HttpMessageNotWritableException { //使用Jackson的ObjectMapper将Java对象转换成Json String ObjectMapper mapper = new ObjectMapper(); String json = mapper.writeValueAsString(object); //加密 String result = AESUtil.encrypt(json); //输出 outputMessage.getBody().write(result.getBytes()); }</code></pre> <h3>基于注解的敏感词过滤功能</h3> <p>项目需要对用户发布的内容进行过滤,将其中的敏感词替换为 * 等特殊字符。大部分Web项目在处理这方面需求时都会选择过滤器( Filter ),在过滤器中将 Request 包上一层 Wrapper ,并重写其 getParameter 等方法,例如:</p> <pre> <code class="language-java">public class SafeTextRequestWrapper extends HttpServletRequestWrapper { public SafeTextRequestWrapper(HttpServletRequest req) { super(req); } @Override public Map<String, String[]> getParameterMap() { Map<String, String[]> paramMap = super.getParameterMap(); for (String[] values : paramMap.values()) { for (int i = 0; i < values.length; i++) { values[i] = SensitiveUtil.filter(values[i]); } } return paramMap ; } @Override public String getParameter(String name) { return SensitiveUtil.filter(super.getParameter(name)); } } public class SafeTextFilter implements Filter { @Override public void init(FilterConfig filterConfig) throws ServletException { } @Override public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain) throws IOException, ServletException { SafeTextRequestWrapper safeTextRequestWrapper = new SafeTextRequestWrapper((HttpServletRequest) request); chain.doFilter(safeTextRequestWrapper, response); } @Override public void destroy() { } }</code></pre> <p>但是这样做会有一些明显的问题,比如无法控制具体对哪些信息进行过滤。如果用户注册的邮箱或是密码中也带有 fuck 之类的敏感词,那就属于误伤了。</p> <p>所以改用Spring MVC的Formatter进行拓展,只需要在 @RequestParam 的参数上使用 @SensitiveFormat 注解,Spring MVC就会在注入该属性时自动进行敏感词过滤。既方便又不会误伤,实现方法如下:</p> <p>声明 @SensitiveFormat 注解:</p> <pre> <code class="language-java">@Target({ElementType.FIELD, ElementType.PARAMETER}) @Retention(RetentionPolicy.RUNTIME) @Documented public @interface SensitiveFormat { }</code></pre> <p>创建 SensitiveFormatter 类。实现 Formatter 接口,重写 parse 方法(将接收到的内容转换成对象的方法),在该方法中对接收内容进行过滤:</p> <pre> <code class="language-java">public class SensitiveFormatter implements Formatter<String> { @Override public String parse(String text, Locale locale) throws ParseException { return SensitiveUtil.filter(text); } @Override public String print(String object, Locale locale) { return object; } }</code></pre> <p>创建 SensitiveFormatAnnotationFormatterFactory 类,实现 AnnotationFormatterFactory 接口,将 @SensitiveFormat 与 SensitiveFormatter 绑定:</p> <pre> <code class="language-java">public class SensitiveFormatAnnotationFormatterFactory implements AnnotationFormatterFactory<SensitiveFormat> { @Override public Set<Class<?>> getFieldTypes() { Set<Class<?>> fieldTypes = new HashSet<>(); fieldTypes.add(String.class); return fieldTypes; } @Override public Printer<?> getPrinter(SensitiveFormat annotation, Class<?> fieldType) { return new SensitiveFormatter(); } @Override public Parser<?> getParser(SensitiveFormat annotation, Class<?> fieldType) { return new SensitiveFormatter(); } }</code></pre> <p>最后将 SensitiveFormatAnnotationFormatterFactory 注册到Spring MVC中:</p> <pre> <code class="language-java">@Configuration public class WebMvcConfig extends WebMvcConfigurerAdapter { @Override public void addFormatters(FormatterRegistry registry) { registry.addFormatterForFieldAnnotation(new SensitiveFormatAnnotationFormatterFactory()); super.addFormatters(registry); } }</code></pre> <h3>记录请求的返回内容</h3> <p>这里提供一种比较通用的方法,基于过滤器实现,所以在非Spring MVC的项目也可以使用。</p> <p>首先导入 commons-io :</p> <pre> <code class="language-java"><dependency> <groupId>commons-io</groupId> <artifactId>commons-io</artifactId> <version>2.4</version> </dependency></code></pre> <p>需要用到这个库中的 TeeOutputStream ,这个类可以将一个将内容同时输出到两个分支的输出流,将其封装为 ServletOutputStream :</p> <pre> <code class="language-java">public class TeeServletOutputStream extends ServletOutputStream { private final TeeOutputStream teeOutputStream; public TeeServletOutputStream(OutputStream one, OutputStream two) { this.teeOutputStream = new TeeOutputStream(one, two); } @Override public boolean isReady() { return false; } @Override public void setWriteListener(WriteListener listener) { } @Override public void write(int b) throws IOException { this.teeOutputStream.write(b); } @Override public void flush() throws IOException { super.flush(); this.teeOutputStream.flush(); } @Override public void close() throws IOException { super.close(); this.teeOutputStream.close(); } }</code></pre> <p>然后创建一个过滤器,将原有的 response 的 getOutputStream 方法重写:</p> <pre> <code class="language-java">public class LoggingFilter implements Filter { private static final Logger LOGGER = LoggerFactory.getLogger(LoggingFilter.class); @Override public void init(FilterConfig filterConfig) throws ServletException { } public void doFilter(ServletRequest request, final ServletResponse response, FilterChain chain) throws IOException, ServletException { final ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream(); HttpServletResponseWrapper responseWrapper = new HttpServletResponseWrapper((HttpServletResponse) response) { private TeeServletOutputStream teeServletOutputStream; @Override public ServletOutputStream getOutputStream() throws IOException { return new TeeServletOutputStream(super.getOutputStream(), byteArrayOutputStream); } }; chain.doFilter(request, responseWrapper); String responseLog = byteArrayOutputStream.toString(); if (LOGGER.isInfoEnabled() && !StringUtil.isEmpty(responseLog)) { LOGGER.info(responseLog); } } @Override public void destroy() { } }</code></pre> <p>将 super.getOutputStream() 和 ByteArrayOutputStream 分别作为两个分支流,前者会将内容返回给客户端,后者使用 toString 方法即可获得输出内容。</p> <p> </p> <p>来自: <a href="/misc/goto?guid=4959674954918551736" rel="nofollow">http://h2ex.com/1198</a></p> <p> </p>