Last updated: 2022-06-13
Persisting JSON with Spring Data
Basically every modern database system has its own data type to persist JSON. Using this type ensures at least the correct formatting - mostly there are also other advantages such as faster I/O.
As Spring Boot developers we want to access the JSON field without manually handling the mapping between the Java entity and the database. In the best case we work exclusively with the Java classes that represent the JSON - the mapping with Jackson and Hibernate is happening transparently.
Technical backgrounds
Spring Data / Hibernate does not support JSON fields out-of-the-box, but there is a library that provides custom types for the most common database systems: hibernate-types. This is a fantastic open source library, please support it!
Let's assume we want to extend our microservice example and add a PartDetails field to the CarPart table. The JSON should have the following structure.
Example JSON we want to add to the car_part table
First we're adding the required dependency to our pom.xml. We're using Maven as in our initial example - Gradle would be no different.
Adding the required depencency
Next we need the model classes representing our JSON. They're containing some validation constraints as well - that's optional but a good thing to ensure before persisting.
Our new model classes, skipping the getters and setters
With this setup we can extend our CarPart entity in the following way.
CarPart class with annotations for supporting JSON
A custom type definition @TypeDef(name = "json", ...) is registered at the class, which becomes active for the field by annotating it with @Type(type = "json"). In our case (MySQL) the class JsonStringType takes over the transformation of PartDetails to JSON. The value for columnDefinition must be "json".
Now we can directly work with the model classes in our code! The JSON is persisted by the custom type, using ObjectMapper in the background.
Creating the custom type in Bootify
Bootify's Free plan also supports JSON fields - just create your custom model in the Data Objects tab and select your model as a field type at the entity. Depending on the database a certain type and column definition is necessary - this is automatically preselected by Bootify together with all the setup around it.
We begin by clicking Start Project - no registration required. Then we go to the Data Objects tab and first create the object "PartDetailsDimensions" with its three fields.
Adding the PartDetailsDimensions data object
Now we create the object "PartDetails". Since "PartDetailsDimensions" already exists, we can select it directly as a custom type.
Adding the PartDetails
Finally we go back to the "Entities" tab and create the CarPart entity if not already existing. Here we can now add the field "partDetails" with its corresponding type PartDetails.
Adding the new field to our entity
That's it! Reading and writing the partDetails now works like a charm without having to worry about any mappings. Bootify creates proper field types and annotations based on the selected database, and the runnable source code can be downloaded right away!
Start Project
No registration required
Further readings
Map JSON objects with generic Hibernate types
