Introduction

Recently, stumbled on this question on the Hibernate forum, and since I’ve been seeing it before on StackOverflow and bumped into it myself while working with JPA and Hibernate, I decided to turn the answer into an article.

Therefore, in this article, you are going to find out how you can fix the “No Dialect mapping for JDBC type” Hibernate issue.

Domain Model

Considering we have a Book entity that defines a properties attribute which is associated with a JSON column in the database.

The Book entity can be mapped as follows:

@Entity(name = "Book")
@Table(name = "book")
public class Book {

    @Id
    @GeneratedValue
    private Long id;

    @NaturalId
    private String isbn;

    @Type(JsonType.class)
    @Column(columnDefinition = "jsonb")
    private JsonNode properties;

    //Getters and setters omitted for brevity
}

The JsonType is provided by the Hypersistence Utils project, so if you want to persist JSON properties, you don’t need to write your own Hibernate Types.

Just add the Hypersistence Utils dependency to your project and map the JSON properties accordingly.

Persisting and fetching the Book entity

Now, let’s assume we have added the following Book entity in our database:

Book book = new Book();
book.setIsbn("978-9730228236");
book.setProperties(
    JacksonUtil.toJsonNode("""
        {
           "title": "High-Performance Java Persistence",
           "author": "Vlad Mihalcea",
           "publisher": "Amazon",
           "price": 44.99
        }
        """
    )
);

entityManager.persist(book);

When persisting the Book entity, Hibernate will issue the proper SQL INSERT statement:

INSERT INTO book (
    isbn, 
    properties, 
    id
) 
VALUES (
    '978-9730228236', 
    {
        "title":"High-Performance Java Persistence",
        "author":
        "Vlad Mihalcea",
        "publisher":"Amazon",
        "price":44.99
    }, 
    1
)

Now, we fetching the Book entity by its natural identifier, we can see that the properties JSON attribute is fetched as a JsonNode:

Book book = entityManager
.unwrap(Session.class)
.bySimpleNaturalId(Book.class)
.load("978-9730228236");

assertEquals(
    "High-Performance Java Persistence", 
    book.getProperties().get("title").asText()
);

Fetching the JSON attribute using JPQL

Now, if we want to fetch the properties entity attribute using JPQL, we can execute the following query:

JsonNode properties = entityManager.createQuery("""
    select b.properties
    from Book b
    where b.isbn = :isbn
    """, JsonNode.class)
.setParameter("isbn", "978-9730228236")
.getSingleResult();

assertEquals(
    "High-Performance Java Persistence", 
    properties.get("title").asText()
);

And everything works properly since the JPQL query is parsed, and the underlying Hibernate Type that handles the properties attribute is going to be known when building the result.

Fetching the JSON attribute using native SQL

However, if we try to do the same using a native SQL query:

JsonNode properties = (JsonNode) entityManager.createNativeQuery("""
    SELECT properties
    FROM book
    WHERE isbn = :isbn
    """)
.setParameter("isbn", "978-9730228236")
.getSingleResult();

assertEquals(
    "High-Performance Java Persistence",
    properties.get("title").asText()
);

Hibernate will throw the following MappingException:

javax.persistence.PersistenceException: org.hibernate.MappingException: No Dialect mapping for JDBC type: 1111

The 1111 JDBC type corresponds to Types.OTHER which is what the PostgreSQL JDBC Driver uses for jsonb column types.

Mapping Types.OTHER to JsonNodeBinaryType

There are several ways you can address this issue. You can register a Hibernate Type to handle the JDBC Types.OTHER either globally or on a per-query basis.

Mapping the JDBC Types.OTHER to JsonNodeBinaryType at the Dialect level

You can map a given JDBC Type code to a Hibernate Type using the database-specific Dialect.

Therefore, for PostgreSQL, we could define a PostgreSQL10JsonDialect that looks as follows:

public class PostgreSQL10JsonDialect
        extends PostgreSQL10Dialect {
 
    public PostgreSQL10JsonDialect() {
        super();
        this.registerHibernateType(
            Types.OTHER, JsonNodeBinaryType.class.getName()
        );
    }
}

And if we provide the custom PostgreSQL10JsonDialect via the hibernate.dialect configuration property:

<property 
    name="hibernate.dialect"
    value="com.vladmihalcea.book.hpjp.hibernate.type.json.PostgreSQL10JsonDialect"
/>

The native SQL query will run just fine.

Mapping the JDBC Types.OTHER to JsonNodeBinaryType at the NativeQuery level

Another option is to provide Hibernate Type associated with the current JDBC ResultSet:

JsonNode properties = (JsonNode) entityManager.createNativeQuery("""
    SELECT properties
    FROM book
    WHERE isbn = :isbn
    """)
.setParameter("isbn", "978-9730228236")
.unwrap(org.hibernate.query.NativeQuery.class)
.addScalar("properties", JsonNodeBinaryType.INSTANCE)
.getSingleResult();

assertEquals(
    "High-Performance Java Persistence",
    properties.get("title").asText()
);

Notice the addScalar method call, which provides the Hibernate Type to be used when handling the Types.Other JDBC column type.

That’s it!

I'm running an online workshop on the 11th of October about High-Performance SQL.

If you enjoyed this article, I bet you are going to love my Book and Video Courses as well.

Conclusion

Handling the No Dialect mapping for JDBC type issue is not very complicated, and it can be done either globally or on a per-query basis.

Insert details about how the information is going to be processed

DOWNLOAD NOW