GenericJackson2JsonRedisSerializer 忽略类和属性

Question

我正在尝试将 ClassA 序列化为 JSON，然后将其反序列化并将其映射到 ClassB 仅在类似属性上，忽略大量检查。假设我有

public class First{
   private String attribute1;
   private String attribute2;
   private String attribute3;
}

public class Second{
   private Int attribute1;
   private String attribute2;
   private String attribute4;
}

我期待的是映射Second 的对象

{
   "attribute2": "value2"
}

有可能吗？我需要这样做，因为我在一些微服务之间有一个共享的 redis 缓存键。由于设计的变化，存储在redis中的类可能会有不同的定义，并且检查和更新每个微服务的代码是不可行的。谢谢

Answer 1

没有直接的方法可以告诉GenericJackson2JsonRedisSerializer 忽略某些字段并将类A 转换为B，您可以实施任何您想要的反序列化策略。

一个简单的例子是，当你想进行类型转换时注册你的映射和可忽略的字段。

// Adapted from spring data redis
public class RqueueRedisSerDes implements RedisSerializer<Object> {
    private ObjectMapper mapper;

    @AllArgsConstructor
    @Getter
    class Dataum {
      Class<?> tgtClass;
      String[] ignorableProperties;
    }

    private Map<Class<?>, Dataum> classMap = new ConcurrentHashMap<>();

    RqueueRedisSerDes() {
      this.mapper = new ObjectMapper();
      this.mapper =
          mapper.registerModule(new SimpleModule().addSerializer(new NullValueSerializer()));
      this.mapper = mapper.configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false);
      this.mapper = mapper.enableDefaultTyping(DefaultTyping.NON_FINAL, As.PROPERTY);
    }

    public void addClassMap(Class<?> source, Class<?> tgt, String[] ignorableProperties) {
      classMap.put(source, new Dataum(tgt, ignorableProperties));
    }

    @Override
    public byte[] serialize(Object source) throws SerializationException {
      if (source == null) {
        return SerializationUtils.EMPTY_ARRAY;
      }
      try {
        return mapper.writeValueAsBytes(source);
      } catch (JsonProcessingException e) {
        throw new SerializationException("Could not write JSON: " + e.getMessage(), e);
      }
    }

    @Override
    public Object deserialize(byte[] source) throws SerializationException {
      if (SerializationUtils.isEmpty(source)) {
        return null;
      }
      try {
        Object object = mapper.readValue(source, Object.class);
        for (Entry<Class<?>, Dataum> entry : classMap.entrySet()) {
          if (ClassUtils.isAssignable(entry.getKey(), object.getClass())) {
            Dataum dataum = entry.getValue();
            Object tgt = dataum.getTgtClass().newInstance();
            BeanUtils.copyProperties(object, tgt, dataum.getIgnorableProperties());
            return tgt;
          }
        }
        return object;
      } catch (Exception ex) {
        throw new SerializationException("Could not read JSON: " + ex.getMessage(), ex);
      }
    }

    private static class NullValueSerializer extends StdSerializer<NullValue> {

      private static final long serialVersionUID = 211020517180777825L;
      private final String classIdentifier;

      NullValueSerializer() {
        super(NullValue.class);
        this.classIdentifier = "@class";
      }

      @Override
      public void serialize(
          NullValue value, JsonGenerator jsonGenerator, SerializerProvider provider)
          throws IOException {
        jsonGenerator.writeStartObject();
        jsonGenerator.writeStringField(classIdentifier, NullValue.class.getName());
        jsonGenerator.writeEndObject();
      }
    }
  }

定义一个实现RedisSerializer<Object>的类，在RedisConnectionFactory中使用这个类来序列化/反序列化值。

class SerializerTest{      
  @Data
  @AllArgsConstructor
  @NoArgsConstructor
  public static class First {
    private String attribute1;
    private String attribute2;
    private String attribute3;
  }
  @Data
  @ToString
  public static class Second {
    private Integer attribute1;
    private String attribute2;
    private String attribute4;
  }

  public static void main(String[] args) {
    RqueueRedisSerDes serDes = new RqueueRedisSerDes();
    // ignore attribute1 due to different type
    serDes.addClassMap(First.class, Second.class, new String[]{"attribute1"});
    First first = new First("1", "2", "3");
    byte[] out = serDes.serialize(first);
    Second second = (Second) serDes.deserialize(out);
    System.out.println(second);
  }
}

我刚刚修改了我的 Repo Rqueue 中的代码