@@ResponseBody 过滤POJO属性

一、问题的提出。 

项目使用Spring MVC框架,并用jackson库处理JSON和POJO的转换。在POJO转化成JSON时,希望动态的过滤掉对象的某些属性。所谓动态,是指的运行时,不同的controler方法可以针对同一POJO过滤掉不同的属性。 

以下是一个Controler方法的定义,使用@ResponseBody把获得的对象列表写入响应的输出流(当然,必须配置jackson的MappingJacksonHttpMessageConverter,来完成对象的序列化)

1
2
3
4
5
6
7
8
@RequestMapping (params = "method=getAllBmForList" )
@ResponseBody
public List<DepartGenInfo> getAllBmForList(HttpServletRequest request,
         HttpServletResponse response) throws Exception {
     
     BmDto dto = bmglService.getAllBm();
     return dto.getBmList();
}

POJO定义

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
public class DepartGenInfo implements java.io.Serializable {
 
      private String depid;
      private String name;
      private Company company;
 
      //getter...
      //setter...
}
 
public class Company  {
 
      private String comid;
      private String name;
<pre name= "code" class = "java" >      //getter...
      //setter...
}

我希望在getAllBmForList返回时,过滤掉DepartGenInfo的name属性,以及company的comid属性。 

jackson支持@JsonIgnore和@JsonIgnoreProperties注解,但是无法实现动态过滤。jackson给出了几种动态过滤的办法,我选择使用annotation mixin 


•JSON View 
•JSON Filter 
•Annotation Mixin 
二、使用annotation mixin动态过滤

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
@RequestMapping (params = "method=getAllBmForList" )
public void getAllBmForList(HttpServletRequest request,
         HttpServletResponse response) throws Exception {
     
     BmDto dto = bmglService.getAllBm();
    
     ObjectMapper mapper = new ObjectMapper();
     SerializationConfig serializationConfig = mapper.getSerializationConfig();
     serializationConfig.addMixInAnnotations(DepartGenInfo. class ,
       DepartGenInfoFilter. class );
 
     serializationConfig.addMixInAnnotations(Company. class ,
       CompanyFilter. class );
     
     mapper.writeValue(response.getOutputStream(),dto.getBmList());
     return ;
}

DepartGenInfoFilter的定义如下:

@JsonIgnoreProperties (value={ "name" }) //希望动态过滤掉的属性
public interface DepartGenInfoFilter {
}<br><span> //CompanyFilter的定义如下: </span>
 

这个实现方法看起来非常不简洁,需要在动态过滤的时候写不少代码,而且也改变了@ResponseBody的运行方式,失去了REST风格,因此考虑到使用AOP来进行处理。 

二、最终解决方案 

先看下我想达到的目标,通过自定义注解的方式来控制动态过滤。

@XunerJsonFilters (value={ @XunerJsonFilter (mixin=DepartGenInfoFilter. class , target=DepartGenInfo. class )
             , @XunerJsonFilter (mixin=CompanyFilter. class , target=Company. class )})
     @RequestMapping (params = "method=getAllBmForList" )
     @ResponseBody
     public List getAllBmForList(HttpServletRequest request,
             HttpServletResponse response) throws Exception {
         
         BmDto dto = bmglService.getAllBm();
return dto.getBmList();
     }

@XunerJsonFilters和@XunerJsonFilter是我定义的注解。@XunerJsonFilters是@XunerJsonFilter的集合,@XunerJsonFilter定义了混合的模板以及目标类。

1
2
3
4
5
6
7
8
9
@Retention (RetentionPolicy.RUNTIME)
public @interface XunerJsonFilters {
     XunerJsonFilter[] value();
}
@Retention (RetentionPolicy.RUNTIME)
public @interface XunerJsonFilter {
   Class<?> mixin() default Object. class ;
   Class<?> target() default Object. class ;
}

当然,只是定义注解并没有什么意义。重要的是如何根据自定义的注解进行处理。我定义了一个AOP Advice如下:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
public class XunerJsonFilterAdvice {
 
     public Object doAround(ProceedingJoinPoint pjp) throws Throwable {
         MethodSignature msig = (MethodSignature) pjp.getSignature();
         XunerJsonFilter annotation = msig.getMethod().getAnnotation(
                 XunerJsonFilter. class );
         XunerJsonFilters annotations = msig.getMethod().getAnnotation(
                 XunerJsonFilters. class );
 
         if (annotation == null && annotations == null ) {
             return pjp.proceed();
         }
 
         ObjectMapper mapper = new ObjectMapper();
         if (annotation != null ) {
             Class<?> mixin = annotation.mixin();
             Class<?> target = annotation.target();
             
             if (target != null ) {
                 mapper.getSerializationConfig().addMixInAnnotations(target,
                         mixin);
             } else {
                 mapper.getSerializationConfig().addMixInAnnotations(
                         msig.getMethod().getReturnType(), mixin);
             }
         }
         
         if (annotations != null ) {
             XunerJsonFilter[] filters= annotations.value();
             for (XunerJsonFilter filter :filters){
                 Class<?> mixin = filter.mixin();
                 Class<?> target = filter.target();
                 
                 if (target != null ) {
                     mapper.getSerializationConfig().addMixInAnnotations(target,
                             mixin);
                 } else {
                     mapper.getSerializationConfig().addMixInAnnotations(
                             msig.getMethod().getReturnType(), mixin);
                 }
             }
             
         }
         
 
         try {
             mapper.writeValue(WebContext.getInstance().getResponse()
                     .getOutputStream(), pjp.proceed());
         } catch (Exception ex) {
             throw new RuntimeException(ex);
         }
         return null ;
     }
 
}

其中pointcut的expression能够匹配到目标类的方法。 

在doAround方法中,需要获得当前引用的HttpResponse对象,因此使用以下方法解决: 

创建一个WebContext工具类:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
public class WebContext {
 
     private static ThreadLocal<WebContext> tlv = new ThreadLocal<WebContext>();
     private HttpServletRequest request;
     private HttpServletResponse response;
     private ServletContext servletContext;
 
     protected WebContext() {
     }
 
     public HttpServletRequest getRequest() {
         return request;
     }
 
     public void setRequest(HttpServletRequest request) {
         this .request = request;
     }
 
     public HttpServletResponse getResponse() {
         return response;
     }
 
     public void setResponse(HttpServletResponse response) {
         this .response = response;
     }
 
     public ServletContext getServletContext() {
         return servletContext;
     }
 
     public void setServletContext(ServletContext servletContext) {
         this .servletContext = servletContext;
     }
 
     private WebContext(HttpServletRequest request,
             HttpServletResponse response, ServletContext servletContext) {
         this .request = request;
         this .response = response;
         this .servletContext = servletContext;
     }
 
     public static WebContext getInstance() {
         return tlv.get();
     }
 
     public static void create(HttpServletRequest request,
             HttpServletResponse response, ServletContext servletContext) {
         WebContext wc = new WebContext(request, response, servletContext);
         tlv.set(wc);
     }
 
     public static void clear() {
         tlv.set( null );
     }
}

别忘了在web.xml中增加这个filter。 




OK,It is all。 




四、总结 

设计的一些要点: 

1、要便于程序员使用。程序员根据业务逻辑需要过滤字段时,只需要定义个"Filter“,然后使用注解引入该Filter。 

2、引入AOP来保持原来的REST风格。对于项目遗留的代码,不需要进行大幅度的修改,只需要增加注解来增加对过滤字段的支持。 

仍需解决的问题: 

按照目前的设计,定义的Filter不支持继承,每一种动态字段的业务需求就会产生一个Filter类,当类数量很多时,不便于管理。 




五、参考资料 

http://www.cowtowncoder.com/blog/archives/cat_json.html 

http://www.jroller.com/RickHigh/entry/filtering_json_feeds_from_spring

你可能感兴趣的:(@@ResponseBody 过滤POJO属性)