Very often, enumerations are stored in SQL databases as
char(1) columns. However,
JPA annotation
Enumerated allows mapping from Java
Enum type into
javax.persistence.EnumType.STRING (i.e. SQL type
VARCHAR) or
javax.persistence.EnumType.ORDINAL (i.e. SQL type
INTERGER) only.
I have read a few rather complicated advices, how to make a mapping from general Java Enum type into SQL
char(1) in
Hibernate. After a while, I have found out, that when all constant names in the Enum have length just character, one can make an easy JPA mapping:
@Basic(optional = false)
@Enumerated(EnumType.STRING)
@Column(nullable = false, columnDefinition = "char(1) default 'A'")
private FruitType fruit_type = FruitType.A;
So, the
FruitType may be:
public enum FruitType {
/** Apples */
A,
/** Apricots */
B,
/** Cherries */
C,
/** Pears */
P;
}
Well, it has a disadvantage, that "apricots" are represented by
FruitType.B. But all modern IDE (like Eclipse) shows Javadoc when the developer types the code. Beside the simplicity, the other great advantage is, that the application developer
knows that the "apricots" are represented by 'B' in the database. So, it is easier for the developer to write JPQL/HQL/SQL queries. When you a complicated JPA/Hibernate mapping from some type
FruitTypeLong.APPRICOT to the 'B' in database, then it will be just confusing after some time to write JPQL/HQL/SQL queries.
