如何配置杰克逊对象映射器以将 Foo 类型的所有字段反序列化为子类的 Foo 实例？

Question

我正在反序列化一个大的json值。深度嵌套在该值中的是一个json对象，如下所示:

{
  "fieldOne": "valueOne",
  "fieldTwo": {
    "innerField": "innerValue"
  }
}

我正在使用JacksonObjectMapper将大的json值反序列化为第3方类。在该第3方类中深度嵌套的是另一个第3方类：

public class DeepThirdPartyClass {
    public String fieldOne;
}

不幸的是，它缺少字段两个属性。我可以创建自己的类来添加缺少的字段：

public class MyClass extends DeepThirdPartyClass {
    public MySubObject fieldTwo;
}

如何配置jackson，以便每当它试图将值反序列化到< code>DeepThirdPartyClass时，它反序列化到< code>MyClass？

Answer 1

我修改了@olga-khylkouskaya的解决方案，以适应我的问题:

@Test
public void newDeserializer() throws Exception {
    ObjectMapper objectMapper = new ObjectMapper();
    SimpleModule module = new SimpleModule("DeepThirdPartyClass subclass override", new Version(1, 0, 0, "FINAL", "com.example", "deep-third-party-class-override"));
    module.addDeserializer(DeepThirdPartyClass.class, new JsonDeserializer<DeepThirdPartyClass>() {
        @Override
        public DeepThirdPartyClass deserialize(JsonParser p, DeserializationContext ctxt) throws IOException, JsonProcessingException {
            return p.readValueAs(MyClass.class);
        }
    });
    objectMapper.registerModule(module);
    objectMapper.configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false);

    String json = "{\n" +
            "  \"middle\": {\n" +
            "    \"fieldOne\": \"valueOne\",\n" +
            "    \"fieldTwo\": {\n" +
            "      \"fieldThree\": \"valueThree\"\n" +
            "    }\n" +
            "  }\n" +
            "}\n";

    ThirdPartyClass thirdPartyClass = objectMapper.readValue(json, ThirdPartyClass.class);
}

public class ThirdPartyClass {
    public DeepThirdPartyClass middle;
}

public class InnerClass {
    public String fieldThree;
}

Answer 2

当我必须过滤所有字符串值中的任何不允许的字符时，我也有类似的要求。创建对象映射器：

public class CustomObjectMapper extends ObjectMapper {
    public CustomObjectMapper() {
        super();

        SimpleModule module = new SimpleModule("HTML XSS Serializer", new Version(1, 0, 0, "FINAL", "com.crowdoptic", "web"));
        module.addSerializer(String.class, new JsonHtmlXssSerializer());
        module.addDeserializer(String.class, new JsonHtmlXssDeserializer());

        this.registerModule(module);
    }
}

public class JsonHtmlXssDeserializer extends StdScalarDeserializer<String> {
    private static final Logger LOG = LogManager.getLogger(JsonHtmlXssDeserializer.class);

    public JsonHtmlXssDeserializer() { super(String.class); }

    @Override
    public String deserialize(JsonParser p, DeserializationContext ctxt) throws IOException {
        String value = StringDeserializer.instance.deserialize(p, ctxt);
        LOG.trace("in deserialize for value: " + value);
        String encodedValue = StringEscapeUtils.escapeHtml4(value);
        return encodedValue;
    }
    @Override
    public String deserializeWithType(JsonParser jp, DeserializationContext ctxt, TypeDeserializer typeDeserializer) throws IOException {
        return StringDeserializer.instance.deserializeWithType(jp, ctxt, typeDeserializer);
    }

    @Override
    public boolean isCachable() { return StringDeserializer.instance.isCachable(); }
}

在您的情况下，您可以注册您的类反序列化器，调用对象反序列化器的超级方法。然后，不返回DeepThirdPartyClass，而是创建MyClass的对象，从DeepThirdPartyClass中设置字段一并添加第二个字段。有关实现细节和可用属性，请参见StringDeserializer和其他内容。

如果有帮助，请告诉我。