ObjectMapper 从具有特定名称的字段创建列表

Question

我有一个 JSON，我将其解析为一个 Result 对象列表。此对象包含 5 个字符串字段，称为 photo1、photo2 等（基于 JSON）是否可以将它们直接读入 List 字段？

JSON 喜欢：

{
    "ErrorMessage": null,
    "Result": [{
        "id": "462290",
        "name_English": "name in english",
        "name_Local": "külföldiül a név",
        "zipcode": "5463",
        "photo1": "dfglkj.com/blabla",
        "photo2": "dfglkj.com/blabla",
        "photo3": "dfglkj.com/blabla",
        "photo4": "dfglkj.com/blabla",
        "photo5": "dfglkj.com/blabla"
    }]
}

还有我的对象：

static final class ApiResponse
    {
        public String ErrorMessage;
        public List<Result> Result = new ArrayList<Result>();
    }

 static final class Result
    {
        public String id;
        public String name_English;
        public String name_Local;
        public List<String> photos;
        public String zipcode;
    }

我有一个 ObjectMapper：

private static ObjectMapper newObjectMapper()
        {
            final ObjectMapper om =
                new ObjectMapper()  //
                .configure(DeserializationConfig.Feature.FAIL_ON_UNKNOWN_PROPERTIES, false)  //
                .configure(JsonParser.Feature.CANONICALIZE_FIELD_NAMES, true);

            om.registerSubtypes(ApiResponse.class);

            return om;
        }

在解析器中：

final ApiResponse ret = OM.readValue(inputStream, ApiResponse.class);

Answer 1

根据 Gaetano 的回答，我写了一个更通用的解决方案（任意数量的照片，任意数量的其他字段

@Override public List<Result> deserialize(JsonParser jp, DeserializationContext ctxt) throws IOException, JsonProcessingException
        {
            JsonNode node = jp.getCodec().readTree(jp);
            List<Result> listResult = new ArrayList<Result>();

            for (JsonNode interNode : node)
            {
                Result result = new Result();

                for (int i = 1; i < 30; i++)
                {
                    if (interNode.get("photo" + i) != null)
                    {
                        result.photos.add(interNode.get("photo" + i).getTextValue());
                    }
                    else
                    {
                        break;
                    }
                }

                for (Field field : result.getClass().getDeclaredFields())
                {
                    if (interNode.get(field.getName()) != null)
                    {
                        try
                        {
                            field.set(result, interNode.get(field.getName()).getTextValue());
                        }
                        catch (IllegalAccessException e)
                        {
                            e.printStackTrace();
                        }
                    }
                }

                listResult.add(result);
            }

            return listResult;
        }

Answer 2

我发现的最佳解决方案是为您的 ObjectMapper 定义一个 Custom Deserializer。

您可以告诉 JACKSON 使用自定义反序列化器，并对 ApiResponse 类上的“结果”对象列表进行特定注释：

public class ApiResponse {

    private String errorMessage;

    @JsonDeserialize(using = CustomDeserializer.class)
    private List<Result> result = new ArrayList<Result>();

    public String getErrorMessage() {
        return errorMessage;
    }

    public void setErrorMessage(String arg) {
        errorMessage = arg;
    }

    public List<Result> getResult() {
        if(result==null)
        {
            result = new ArrayList<Result>();
        }
        return result;
    }

    public void setResult(List<Result> arg) {
        result = arg;
    }
}

public class Result {
     private String id;
     private String name_English;
     private String name_Local;
     private List<String> photos;
     private String zipcode;
    public String getId() {
        return id;
    }
    public void setId(String id) {
        this.id = id;
    }
    public String getName_English() {
        return name_English;
    }
    public void setName_English(String name_English) {
        this.name_English = name_English;
    }
    public String getName_Local() {
        return name_Local;
    }
    public void setName_Local(String name_Local) {
        this.name_Local = name_Local;
    }
    public List<String> getPhotos() {
        if(photos == null)
        {
            photos = new ArrayList<String>();
        }
        return photos;
    }
    public void setPhotos(List<String> photos) {
        this.photos = photos;
    }
    public String getZipcode() {
        return zipcode;
    }
    public void setZipcode(String zipcode) {
        this.zipcode = zipcode;
    }
}

请注意java命名约定：https://docs.oracle.com/javase/tutorial/java/nutsandbolts/variables.html

自定义反序列化器将如下所示：

public class CustomDeserializer extends StdDeserializer<List<Result>> {

private static final long serialVersionUID = -3483096770025118080L;

public CustomDeserializer() {
    this(null);
}

public CustomDeserializer(Class<?> vc) {
    super(vc);
}

@Override
public List<Result> deserialize(JsonParser jp, DeserializationContext ctxt)
        throws IOException, JsonProcessingException {
    JsonNode node = jp.getCodec().readTree(jp);

    List<Result> listResult = new ArrayList<Result>();
    for (JsonNode interNode : node) {

        Result result = new Result();           

        if (interNode.get("id") != null) {
            result.setId(interNode.get("id").asText());
        }
        if (interNode.get("name_English") != null) {
            result.setName_English(interNode.get("name_English").asText());
        }
        if (interNode.get("name_Local")!= null) {
            result.setName_Local(interNode.get("name_Local").asText());
        }
        if (interNode.get("zipcode") !=null ) {
            result.setZipcode(interNode.get("zipcode").asText());
        }

        // photo array
        if (interNode.get("photo1") != null) {
            result.getPhotos().add(interNode.get("photo1").asText());
        }
        if (interNode.get("photo2") != null) {
            result.getPhotos().add(interNode.get("photo2").asText());
        }
        if (interNode.get("photo3") != null) {
            result.getPhotos().add(interNode.get("photo3").asText());
        }
        if (interNode.get("photo4") != null) {
            result.getPhotos().add(interNode.get("photo4").asText());
        }
        if (interNode.get("photo5") != null) {
            result.getPhotos().add(interNode.get("photo5").asText());
        }

        listResult.add(result);
    }

    return listResult;
}

Answer 3

如果您可以更改 POJO 代码，那么最简单的选择是使用 @JsonAnySetter 和 @JsonAnyGetter 并在那里编写自定义代码。（如建议的 @Oleg）但不要忘记 @JsonIgnore List<String> photos。

@Data // lombok.Data; for getters and setters
public static final class Result {
    public String id;
    public String name_English;
    public String name_Local;
    @JsonIgnore
    public List<String> photos = new ArrayList<>();
    public String zipcode;

    @JsonAnySetter
    public void setOther(String key, String value){
        photos.add(value);
    }
    @JsonAnyGetter
    public Map<String,String> getOther(){
        Map<String,String> map = new HashMap<>();
        for (int i = 0;i<photos.size();i++)
            map.put("photo" + i + 1, photos.get(i));
        return map;
    }
}

但是如果你不能改变 POJO 代码，你将不得不编写自定义的序列化器和反序列化器