0wl
0wl

Reputation: 876

Custom deserializer for any list in Jackson

I have a problem with wrong objects in lists. For instance I've a JSON model:

{
  "items": [
    {
      "id": 1,
      "name": "Item1"
    },
    {
      "id": 2,
      "name": "Item2"
    },
    {
      "id": [],
      "name": "Item3"
    }
  ]
}

and two POJO

data class BadList(val items: List<BadItem>)

data class BadItem(val id: Int, val name: String)

Of course, when the parser stumbles upon a third element I get the exception

com.fasterxml.jackson.databind.JsonMappingException: Can not deserialize instance of java.lang.Integer out of START_ARRAY token
 at [Source: {"items":[{"id":1,"name":"Item1"},{"id":2,"name":"Item2"},{"id":[],"name":"Item3"}]}; line: 1, column: 19] (through reference chain: my.package.BadList["items"]->java.util.ArrayList[2]->my.package.BadItem["id"])

Who knows how to get around this? I want to skip that wrong item.

Upvotes: 0

Views: 4532

Answers (2)

Sven Hasselbach
Sven Hasselbach

Reputation: 10485

You can use a "HidableSerializer" for this and check the data during serialization

1. Create a IHidable interface

The interface has a isHidden method which is called during serialization

package ch.hasselba.jackson.test;

public interface IHidable {
    public boolean isHidden();
}

2. Change your BadItem class

Add the interface and change the setter of id. When property id is deserialized, it is tested if it is an Integer. If not, the item is marked as bad.

package ch.hasselba.jackson.test;

import com.fasterxml.jackson.annotation.JsonIgnoreProperties;

@JsonIgnoreProperties( {"hidden"} ) 
public class BadItem implements IHidable{
    private Integer id;
    public String name;
    private boolean isBadItem;
    public Integer getId(){
        return id;
    }
    public void setId(Object value){
        if( value instanceof Integer ){
            this.id = (Integer) value;
        }else{
            this.isBadItem = true;
        }
    }
    public boolean isHidden() {
        return isBadItem;
    }
}

3. Create a HidableSerializer

package ch.hasselba.jackson.test;

import java.io.IOException;

import com.fasterxml.jackson.core.JsonGenerator;
import com.fasterxml.jackson.core.JsonProcessingException;
import com.fasterxml.jackson.databind.JsonSerializer;
import com.fasterxml.jackson.databind.SerializerProvider;
import com.fasterxml.jackson.databind.ser.std.StdSerializer;


@SuppressWarnings("serial")
public class HidableSerializer<T> extends StdSerializer<T> {

    private JsonSerializer<T> defaultSerializer;

    protected HidableSerializer(Class<T> t) {
        super(t);
    }

    public JsonSerializer<T> getDefaultSerializer() {
        return defaultSerializer;
    }

    public void setDefaultSerializer(JsonSerializer<T> defaultSerializer) {
        this.defaultSerializer = defaultSerializer;
    }

    @Override
    public void serialize(T value, JsonGenerator jgen, SerializerProvider provider)
            throws IOException, JsonProcessingException {

        if( value instanceof IHidable ){
            IHidable hidableValue = (IHidable) value;
            if( hidableValue.isHidden() )
                return;
        }

        defaultSerializer.serialize(value, jgen, provider);
    }

}

4. Register the HidableSerializer and that's it

package ch.hasselba.jackson.test;

import com.fasterxml.jackson.annotation.JsonInclude.Include;
import com.fasterxml.jackson.databind.BeanDescription;
import com.fasterxml.jackson.databind.JsonSerializer;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.fasterxml.jackson.databind.SerializationConfig;
import com.fasterxml.jackson.databind.module.SimpleModule;
import com.fasterxml.jackson.databind.ser.BeanSerializerModifier;

public class Demo {

    @SuppressWarnings("serial")
    public static void main(String[] args) {

        // register the HidableSerializer
        ObjectMapper mapper = new ObjectMapper();
        mapper.setSerializationInclusion(Include.NON_EMPTY);
        mapper.registerModule(new SimpleModule() {
            @Override
            public void setupModule(SetupContext context) {
                super.setupModule(context);
                context.addBeanSerializerModifier(new BeanSerializerModifier() {
                    @Override
                    public JsonSerializer<?> modifySerializer(
                      SerializationConfig config, BeanDescription desc, JsonSerializer<?> serializer) {
                        if (BadItem.class.isAssignableFrom(desc.getBeanClass())) {
                            HidableSerializer ser = new HidableSerializer(BadItem.class);
                            ser.setDefaultSerializer(serializer);
                            return ser;
                        }
                        return serializer;
                    }
                });
            }
        });

        String content = "{  \"items\": [    {      \"id\": 1,      \"name\": \"Item1\"    },    {      \"id\": 2,      \"name\": \"Item2\"    },    {      \"id\":[],      \"name\": \"Item3\"    }  ]}";

        // build the Object
        BadList test = null;
        try {
            test =  mapper.readValue(content, BadList.class);
        } catch (Exception e) {
            e.printStackTrace();
        }

        // and now convert it back to a String
        String data = null;
        try {
             data = mapper.writeValueAsString(test);
        } catch (Exception e) {
            e.printStackTrace();
        }

        // print the result
        System.out.println( data );
    }

}

When changing the id "[]" to an Integer value, the Item is displayed, otherwise it is empty.

The result:

{"items":[{"id":1,"name":"Item1"},{"id":2,"name":"Item2"}]}

Upvotes: 1

Darshan Mehta
Darshan Mehta

Reputation: 30839

You can write a custom deserializer and implement deserialization logic in it, e.g.:

class ItemIdDeserialiser extends JsonDeserializer<Integer> {

    @Override
    public Integer deserialize(JsonParser p, DeserializationContext ctxt) throws IOException, JsonProcessingException {
        Object value = p.getCurrentValue();
        //Check if it's Integer
        if(value instanceof Integer){
            return (Integer) value;
        }
        return null; //Or return first element if it's a non empty list
    }
}

Once this is done, you can annotate the field with @JsonDeserialise to instruct jackson to use your class, e.g.:

class Item {

    @JsonDeserialize(using = ItemIdDeserialiser.class)
    private Integer id;
}

Update

If you just want to ignore the field in serialization/deserialization then you can annotate it with @JsonIgnore, e.g.

class Item {

        @JsonIgnore
        private Integer id;
    }

Or even better, remove id from pojo and add @JsonIgnoreProperties on the class, e.g.:

@JsonIgnoreProperties(ignoreUnknown = true)
class Item {

}

It will automatically ignore the properties which are present in json but not found in class.

Upvotes: 2

Related Questions