DEV Community

AL
AL

Posted on • Edited on

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;
    }
}
Enter fullscreen mode Exit fullscreen mode

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;
    }
}
Enter fullscreen mode Exit fullscreen mode

ContactDto

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

package com.microservice.crm.fixtures;

import com.fasterxml.jackson.annotation.JsonAutoDetect;
import com.fasterxml.jackson.annotation.JsonFormat;
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;
import com.microservice.crm.annotation.*;
import com.microservice.crm.serializer.MappingValue;
import com.microservice.crm.serializer.MappingValueDeserializer;
import com.microservice.crm.serializer.MappingValueSerializer;

import java.time.LocalDate;

@JsonAutoDetect(
    fieldVisibility = JsonAutoDetect.Visibility.ANY,
    getterVisibility = JsonAutoDetect.Visibility.NONE,
    setterVisibility = JsonAutoDetect.Visibility.NONE,
    isGetterVisibility = JsonAutoDetect.Visibility.NONE
)
@JsonInclude(JsonInclude.Include.NON_NULL)
@JsonIgnoreProperties(ignoreUnknown = true)
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 = "{\"1\": \"MALE\", \"2\": \"FEMALE\", \"6\": \"DIVERS\"}")
    @JsonSerialize(using = MappingValueSerializer.class)
    @JsonDeserialize(using = MappingValueDeserializer.class)
    private MappingValue<String> salutation;

    @JsonProperty("100674")
    @MappingTable(map = "{\"1\": true, \"2\": false}")
    @JsonSerialize(using = MappingValueSerializer.class)
    @JsonDeserialize(using = MappingValueDeserializer.class)
    private MappingValue<Boolean> marketingInformation;

    public String getFirstname() {
        return firstname;
    }

    public void setFirstname(String firstname) {
        this.firstname = firstname;
    }

    public String getLastname() {
        return lastname;
    }

    public void setLastname(String lastname) {
        this.lastname = lastname;
    }

    public String getEmail() {
        return email;
    }

    public void setEmail(String email) {
        this.email = email;
    }

    public LocalDate getBirthday() {
        return birthday;
    }

    public void setBirthday(LocalDate birthday) {
        this.birthday = birthday;
    }

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

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

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

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

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
)
Enter fullscreen mode Exit fullscreen mode

FieldIDs

The FieldIDs can be defined very easy with @JsonProperty.

@JsonProperty("123")
Enter fullscreen mode Exit fullscreen mode

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)
Enter fullscreen mode Exit fullscreen mode

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)
Enter fullscreen mode Exit fullscreen mode

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)
Enter fullscreen mode Exit fullscreen mode

Custom Annotation @MappingTable for the MappingTable

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

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

package com.microservice.crm.annotation;

import java.lang.annotation.*;

@Retention(RetentionPolicy.RUNTIME)
@Target(ElementType.FIELD)
@Documented
public @interface MappingTable {
    String map() default "{}";
}
Enter fullscreen mode Exit fullscreen mode

The @MappingTable is defined as masked JSON (String) at the annotation.

Only the following types are possible for annotations:

  • Primitive types
  • String
  • Enum
  • Class (Class <?>, Class<? extends/super T>)
  • Array of the above (array[] of primitives, enums, String, or Class)
  • Another annotation.

See this conversation on Stackoverflow.

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.

The MappingTableDataReader converts the JSON into a HashMap.

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.microservice.crm.serializer;

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;
import com.microservice.crm.annotation.MappingTableDataReader;

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

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

    private final HashMap<String, ?> data;

    public MappingValueSerializer() {
        this(null);
    }

    public MappingValueSerializer(HashMap<String, ?> data) {
        this.data = data;
    }

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

    @Override
    public JsonSerializer<?> createContextual(SerializerProvider prov, BeanProperty property) {
        return new MappingValueSerializer(
                new MappingTableDataReader().getMap(property)
        );
    }
}
Enter fullscreen mode Exit fullscreen mode

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.microservice.crm.serializer;

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;
import com.fasterxml.jackson.databind.type.SimpleType;
import com.microservice.crm.annotation.MappingTableDataReader;

import java.io.IOException;
import java.lang.reflect.Type;
import java.util.Arrays;
import java.util.HashMap;
import java.util.Map;

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

    private final String[] supportedTypes = {"String", "Boolean"};

    private final HashMap<String, ?> data;

    private final Type type;

    public MappingValueDeserializer() {
        this(null, null);
    }

    public MappingValueDeserializer(HashMap<String, ?> data, Type type) {
        this.data = data;
        this.type = type;
    }

    @Override
    public MappingValue<?> deserialize(JsonParser jsonParser, DeserializationContext deserializationContext) throws IOException {
        String value = jsonParser.getText();
        String simpleName = ((SimpleType) this.type).getBindings().getTypeParameters().get(0).getRawClass().getSimpleName();

        if (Arrays.stream(supportedTypes).noneMatch(simpleName::equalsIgnoreCase)) {
            throw new IOException(String.format("Type \"%s\" is currently not supported", simpleName));
        }

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

    @Override
    public JsonDeserializer<?> createContextual(DeserializationContext ctxt, BeanProperty property) {
        return new MappingValueDeserializer(
                new MappingTableDataReader().getMap(property),
                property.getType()
        );
    }
}
Enter fullscreen mode Exit fullscreen mode

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.microservice.crm.serializer;

import com.fasterxml.jackson.databind.ObjectMapper;
import com.microservice.crm.fixtures.ContactDto;
import net.javacrumbs.jsonunit.core.Option;
import org.junit.jupiter.api.Test;

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

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

public 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.trim())
                .when(Option.IGNORING_ARRAY_ORDER)
                .isEqualTo(json);
    }

    @Test
    void deserialize() throws IOException {
        ContactDto contact = new ObjectMapper().readValue(this.emarsysPayload.trim(), 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());
    }
}
Enter fullscreen mode Exit fullscreen mode

ToDo

The MappingTable is defined as masked JSON to the annotation @MappingTable. This means we cannot use the data in any other context. A HashMap cannot be defined because Annotation does not support HashMap. For example, the solution would be to use ENUM class.

How this can be solved I will look at in a future article.

Links

Full example on github

Updates

  • Change GitHub Repository URL (Sep 6th 2024)

Top comments (0)

Read next

lasserafn profile image

11 API Documentation Best Practices for CI/CD 2024

Lasse Foo-Rafn -

andrewkangg profile image

Why would I use this instead of Traefik for zero-downtime deployment?

Andrew Kang-G -

ajmal_hasan profile image

Seamless Integration of Hugging Face AI Models via API for Any Application

Ajmal Hasan -

trinityyi profile image

REST vs. GraphQL: Choosing the Right API for Your Project

Isabelle M. -