André Laugks

Posted on May 11, 2023 • Edited on Dec 22

Create a custom Jackson JsonSerializer und JsonDeserializer for mapping values

#java #jackson #api #springboot

For my series of articles, I also wanted to see how this requirement to mapping values could be implemented with Jackson.

The first paragraph "The requirements and history" from the first article describes the requirements for Emarsys to rewrite the values for the payload.

The required packages

com.fasterxml.jackson.core:jackson-databin
com.fasterxml.jackson.datatype:jackson-datatype-jsr310

See the pom.xml in the example for the latest versions.

Minimal structure of a custom JsonSerializer and JsonDeserializer

To solve the requirements to map the values for Emarsys, a custom JsonSerializer and JsonDeserializer is needed. I call these MappingValueSerializer and MappingValueDeserializer.

Below is the minimal structure of a custom MappingValueSerializer and MappingValueDeserializer:

@JsonSerialize(using = MappingValueSerializer.class)
@JsonDeserialize(using = MappingValueDeserializer.class)
private String fieldName;

public class MappingValueSerializer extends JsonSerializer<String> {
    @Override
    public void serialize(String value, JsonGenerator jsonGenerator, SerializerProvider serializerProvider) throws IOException {
        jsonGenerator.writeString("serialized: " + value);
    }
}

public class MappingValueDeserializer extends JsonDeserializer<String> {
    @Override
    public String deserialize(JsonParser jsonParser, DeserializationContext deserializationContext) throws IOException  {
        String value = jsonParser.getText();
        return "deserialized: " + value;
    }
}

In ContactDto, the fields salutation and marketingInformation for which values have to be rewritten are defined.

Fields/Direction	serialize (T -> String)	deserialize (String -> T)
salutation	"FEMALE" -> "2"	"2" -> "FEMALE"
marketingInformation	true -> "1"	"1" -> true

For the serialize process it is the FieldValueID (String) and for the deserialize process the type String for salutation and the type Boolean for marketingInformation.

So if you want to do the mapping, you would need a JsonSerializer to write the FieldValueID (String) for salutation and marketingInformation and a JsonDeserializer to set the value for stalutation (String) and marketingInformation (Boolean).

Custom Type

However, I only want to have a JsonDeserializer that can process String, Boolean and in the future other types. For this purpose, I create my own type MappingValue<>. Most importantly, I can transport all types with this custom generics.

package com.microservice.crm.serializer;

public class MappingValue<T> {
    T value;

    public MappingValue(T value) {
        this.value = value;
    }

    public T getValue() {
        return this.value;
    }
}

ContactDto

First of all the complete ContactDto with all fields and annotations. I will explain the individual annotations below.

package com.article.jackson.dto;

import java.time.LocalDate;

import com.article.jackson.annotation.MappingTable;
import com.article.jackson.serializer.MappingValue;
import com.article.jackson.serializer.MappingValueDeserializer;
import com.article.jackson.serializer.MappingValueSerializer;
import com.fasterxml.jackson.annotation.JsonAutoDetect;
import com.fasterxml.jackson.annotation.JsonFormat;
import com.fasterxml.jackson.annotation.JsonIgnoreProperties;
import com.fasterxml.jackson.annotation.JsonInclude;
import com.fasterxml.jackson.annotation.JsonProperty;
import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
import com.fasterxml.jackson.databind.annotation.JsonSerialize;
import com.fasterxml.jackson.datatype.jsr310.deser.LocalDateDeserializer;
import com.fasterxml.jackson.datatype.jsr310.ser.LocalDateSerializer;

@JsonIgnoreProperties(ignoreUnknown = true)
@JsonInclude(JsonInclude.Include.NON_NULL)
@JsonAutoDetect(
    fieldVisibility = JsonAutoDetect.Visibility.ANY,
    getterVisibility = JsonAutoDetect.Visibility.NONE,
    setterVisibility = JsonAutoDetect.Visibility.NONE,
    isGetterVisibility = JsonAutoDetect.Visibility.NONE
)
public class ContactDto {

    @JsonProperty("1")
    private String firstname;

    @JsonProperty("2")
    private String lastname;

    @JsonProperty("3")
    private String email;

    @JsonProperty("4")
    @JsonFormat(pattern = "yyyy-MM-dd")
    @JsonSerialize(using = LocalDateSerializer.class)
    @JsonDeserialize(using = LocalDateDeserializer.class)
    private LocalDate birthday;

    @JsonProperty("46")
    @MappingTable(map = Maps.SALUTATION)
    @JsonSerialize(using = MappingValueSerializer.class)
    @JsonDeserialize(using = MappingValueDeserializer.class)
    private MappingValue<String> salutation;

    @JsonProperty("100674")
    @MappingTable(map = Maps.MARKETING_INFORMATION)
    @JsonSerialize(using = MappingValueSerializer.class)
    @JsonDeserialize(using = MappingValueDeserializer.class)
    private MappingValue<Boolean> marketingInformation;

    // other getter and setter

    public String getSalutation() {
        return salutation.getValue();
    }

    public void setSalutation(String salutation) {
        this.salutation = new MappingValue<>(salutation);
    }

    public Boolean getMarketingInformation() {
        return this.isMarketingInformation();
    }

    public Boolean isMarketingInformation() {
        return marketingInformation.getValue();
    }

    public void setMarketingInformation(Boolean marketingInformation) {
        this.marketingInformation = new MappingValue<>(marketingInformation);
    }
}

Custom Annotation @MappingTable and Enum Constants for the MappingTable

The MappingTable with the FieldValueIDs for salutation and marketingInformation must be available in the MappingValueSerializer and MappingValueDeserializer.

Annotation @MappingTable

For this I create a custom annotation @MappingTable that will be read in the MappingValueSerializer and MappingValueDeserializer.

package com.article.jackson.annotation;

import java.lang.annotation.ElementType;
import java.lang.annotation.Retention;
import java.lang.annotation.RetentionPolicy;
import java.lang.annotation.Target;

import com.article.jackson.dto.Maps;

@Retention(RetentionPolicy.RUNTIME)
@Target(ElementType.FIELD)
public @interface MappingTable {
    Maps map();
}

Enum Constants

package com.article.jackson.dto;

import java.util.Map;

public enum Maps {

    SALUTATION(Map.of("1", "MALE", "2", "FEMALE", "6", "DIVERS")),

    MARKETING_INFORMATION(Map.of("1", true, "2", false));

    private final Map<String, Object> map;

    Maps(Map<String, Object> map) {
        this.map = map;
    }

    public Map<String, Object> getMap() {
        return this.map;
    }
}

Field definitions

The enum constants Maps.SALUTATION and Maps.MARKETING_INFORMATION are referenced in the @MappingTable annotation. The HashMaps are used in the JsonSerializer and JsonDeserializer for bi-directional mapping.

@MappingTable(map = Maps.SALUTATION)
private MappingValue<String> salutation;

@MappingTable(map = Maps.MARKETING_INFORMATION)
private MappingValue<Boolean> marketingInformation;

Reading and writing on the fields

The ObjectManager of Jackson writes and reads on the mutators (setter) and accessor (getter, isser) by default.

For the mutator and accessor of salutation and marketingInformation, however, I would like to define the type String or Boolean.

You can use an annotation to instruct Jackson to read and write only on the fields, so we can use the custom type MappingValue<> internally. The reading and writing process thus takes place on the fields and we can define String and Boolean for the mutator and accessor of salutation and marketingInformation.

@JsonAutoDetect(
    fieldVisibility = JsonAutoDetect.Visibility.ANY,
    getterVisibility = JsonAutoDetect.Visibility.NONE,
    setterVisibility = JsonAutoDetect.Visibility.NONE,
    isGetterVisibility = JsonAutoDetect.Visibility.NONE
)

FieldIDs

The FieldIDs can be defined very easy with @JsonProperty.

@JsonProperty("123")

Define custom JsonSerializer and JsonDeserializer

The custom JsonSerializer (MappingValueSerializer) and JsonDeserializer (MappingValueDeserializer) can be defined with @JsonSerialize and @JsonDeserialize on the field.

@JsonSerialize(using = MappingValueSerializer.class)
@JsonDeserialize(using = MappingValueDeserializer.class)

Skip null values

Fields with null as value should not be serialized. This is because the fields that are sent are also updated. The annotation @JsonInclude can be used for this.

@JsonInclude(JsonInclude.Include.NON_NULL)

Ignore unknown properties

Emarsys always returns all fields for a contact in the response. I want only the fields defined in the ContactDto to be mapped so that no exceptions are thrown. The annotation @JsonIgnoreProperties can be used for this:

@JsonIgnoreProperties(ignoreUnknown = true)

MappingValueSerializer and MappingValueDeserializer

In order for the MappingTable, which is defined at the field, to be read, the interface ContextualSerializer must be implemented for the MappingValueSerializer and the interface ContextualDeserializer for the MappingValueDeserializer.

With createContextual(), access to the property is possible and via BeanProperty the annotation can be fetched and the MappingTable can be read out.

MappingValueSerializer

In the MappingValueSerializer, for example, for salutation "FEMALE" is mapped to "2" and marketingInformation true to "1", which is why the FieldValueID is written with jsonGenerator.writeString().

package com.article.jackson.serializer;

import java.io.IOException;
import java.util.Map;

import com.article.jackson.annotation.MappingTableMapReader;
import com.fasterxml.jackson.core.JsonGenerator;
import com.fasterxml.jackson.databind.BeanProperty;
import com.fasterxml.jackson.databind.JsonSerializer;
import com.fasterxml.jackson.databind.SerializerProvider;
import com.fasterxml.jackson.databind.ser.ContextualSerializer;

public class MappingValueSerializer extends JsonSerializer<MappingValue<?>> implements ContextualSerializer {

    private final Map<String, Object> map;

    public MappingValueSerializer() {
        this(null);
    }

    public MappingValueSerializer(Map<String, Object> map) {
        this.map = map;
    }

    @Override
    public void serialize(MappingValue<?> field, JsonGenerator jsonGenerator, SerializerProvider serializerProvider) throws IOException {
        String fieldValueId = this.map.entrySet()
                .stream()
                .filter(e -> e.getValue().equals(field.getValue()))
                .map(Map.Entry::getKey)
                .findFirst()
                .orElse(null);

        jsonGenerator.writeString(fieldValueId);
    }

    @Override
    public JsonSerializer<?> createContextual(SerializerProvider prov, BeanProperty property) {
        return new MappingValueSerializer(
                new MappingTableMapReader(property).getMap()
        );
    }
}

MappingValueDeserializer

In the MappingValueDeserializer the mapping takes place backwards. Here the FieldValueID for salutation and marketingInformation must be mapped accordingly. For salutation "2" to "FEMALE" (String) and for marketingInformation "1" to true (Boolean).

package com.article.jackson.serializer;

import java.io.IOException;
import java.util.Map;

import com.article.jackson.annotation.MappingTableMapReader;
import com.fasterxml.jackson.core.JsonParser;
import com.fasterxml.jackson.databind.BeanProperty;
import com.fasterxml.jackson.databind.DeserializationContext;
import com.fasterxml.jackson.databind.JsonDeserializer;
import com.fasterxml.jackson.databind.deser.ContextualDeserializer;

public class MappingValueDeserializer extends JsonDeserializer<MappingValue<?>> implements ContextualDeserializer {

    private final Map<String, Object> map;

    public MappingValueDeserializer() {
        this(null);
    }

    public MappingValueDeserializer(Map<String, Object> map) {
        this.map = map;
    }

    @Override
    public MappingValue<?> deserialize(JsonParser jsonParser, DeserializationContext deserializationContext) throws IOException {
        String fieldValue = jsonParser.getText();

        return new MappingValue<>(this.map.entrySet()
                .stream()
                .filter(e -> e.getKey().equals(fieldValue))
                .map(Map.Entry::getValue)
                .findFirst()
                .orElse(null));
    }

    @Override
    public JsonDeserializer<?> createContextual(DeserializationContext ctxt, BeanProperty property) {
        return new MappingValueDeserializer(
                new MappingTableMapReader(property).getMap()
        );
    }
}

MappingTableMapReader

In the MappingTableMapReader class, the enum constant is retrieved from the annotation and is available in the JsonSerializer and JsonDeserializer.

ackage com.article.jackson.annotation;

import java.util.Map;

import com.article.jackson.exception.MappingTableRuntimeException;
import com.fasterxml.jackson.databind.BeanProperty;

public class MappingTableMapReader {

    private final BeanProperty property;

    public MappingTableMapReader(BeanProperty property) {
        this.property = property;
    }

    public Map<String, Object> getMap() {

        MappingTable annotation = property.getAnnotation(MappingTable.class);

        if (annotation == null) {
            throw new MappingTableRuntimeException(
                String.format(
                    "Annotation @MappingTable not set at property %s",
                    this.property.getMember().getFullName()
                )
            );
        }

        return annotation.map().getMap();
    }
}

Functional Test

To check the implementation, we still need a test. To compare the JSON, I use assertThatJson() from the package json-unit-assertj.

package com.article.jackson.serializer;

import java.io.IOException;
import java.time.LocalDate;

import com.article.jackson.dto.ContactDto;
import com.article.jackson.exception.MappingTableRuntimeException;
import com.article.jackson.fixtures.ContactDtoAnnotationNotSet;
import com.fasterxml.jackson.core.JsonProcessingException;
import com.fasterxml.jackson.databind.ObjectMapper;
import net.javacrumbs.jsonunit.core.Option;
import org.junit.jupiter.api.Test;

import static net.javacrumbs.jsonunit.assertj.JsonAssertions.assertThatJson;
import static org.junit.jupiter.api.Assertions.assertEquals;
import static org.junit.jupiter.api.Assertions.assertTrue;

class MappingTableSerializerDeserializerTest {

    String emarsysPayload = """
            {
                "1": "Jane",
                "2": "Doe",
                "3": "jane.doe@example.com",
                "4": "1989-11-09",
                "46": "2",
                "100674": "1"
            }
            """;

    @Test
    void serialize() throws IOException {
        ContactDto contact = new ContactDto();
        contact.setSalutation("FEMALE");
        contact.setFirstname("Jane");
        contact.setLastname("Doe");
        contact.setEmail("jane.doe@example.com");
        contact.setBirthday(LocalDate.of(1989, 11, 9));
        contact.setMarketingInformation(true);
        String json = new ObjectMapper().writeValueAsString(contact);
        assertThatJson(this.emarsysPayload)
                .when(Option.IGNORING_ARRAY_ORDER)
                .isEqualTo(json);
    }

    @Test
    void deserialize() throws IOException {
        ContactDto contact = new ObjectMapper().readValue(this.emarsysPayload, ContactDto.class);
        assertEquals("FEMALE", contact.getSalutation());
        assertEquals("Jane", contact.getFirstname());
        assertEquals("Doe", contact.getLastname());
        assertEquals("jane.doe@example.com", contact.getEmail());
        assertEquals(LocalDate.of(1989, 11, 9), contact.getBirthday());
        assertTrue(contact.getMarketingInformation());
        assertTrue(contact.isMarketingInformation());
    }
}

DEV Community