[sliderrr]

I don’t really know if this is a bug or working as intended, mybe someone can explain this behavior:

If I generate the entity classes with the JPA Reverse Engineering tools form a postgresql database it generates two files, one “default” class named like the table with an embedded id and one Id class with the actual table contents.
Example:
Database:

CREATE TABLE customer
(
  email character varying NOT NULL,
  prefix character varying NOT NULL,
  firstname character varying(30),
  lastname character varying(30),
  ....
)

Some index-fields, but pretty much a plain table.

Now the class looks like this:

customer.java:

@Entity
@Table(name = "customer", schema = "public", uniqueConstraints = @UniqueConstraint(columnNames = "id"))
public class Customer implements java.io.Serializable {

    // Fields

    private CustomerId id;
       ....
    // Property accessors
    @EmbeddedId
    @AttributeOverrides( {
            @AttributeOverride(name = "id", column = @Column(name = "id", unique = true, nullable = false)),
            @AttributeOverride(name = "firstname", column = @Column(firstname = "firstname", nullable = false)),

customerid.java

@Embeddable
public class CustomerId implements java.io.Serializable {

    // Fields

    private Integer id;
    private String firstname;
       ....

Is there a specific reason for that (splitting into two files with EmbeddedId) or am I missing something else?

Thanks in advance.

[Brian Fernandes]

sliderrr,

Yes, this is working as expected. This is because you have a composite primary key (a primary key with more than one column). In such cases the tooling will generate a separate entity for all columns in the primary key and “embed” that key into the primary entity.

This makes it easier for you to reference a particular customer in relationships using a single CustomerId class.

Hope this helps

[sliderrr]

The tool was actually working correctly, the problem was in my case that the primary key was just a id-column with an unique constraint. The values were generated from a nextval-sequence table which caused the tool to compose the id from all columns becaus I could recognize the id.

Thanks for the help 🙂

[Author]

Posts