I'm trying to do a very simple delete operation, but somehow it doesn't work since I updated the DAOs to JpaRepository. Basically it's this:
A a = aRepository.findOne(id);
a.setSomeField("someNewString");
List<B> bList = a.getBs();
bList.clear();
aRepository.saveAndFlush(a);
The field get's updated as expected, but the bList stays unchanged. I've even tried:
A a = aRepository.findOne(id);
a.setSomeField("someNewString");
List<B> bList = a.getBs();
for(B b : bList) {
   bRepository.delete(b);
}
bRepository.flush();
bList.clear();
aRepository.saveAndFlush(a);
Still the same...
Class A looks like this:
@Entity
@Table(name = "A")
public class A implements Serializable {
    private static final long serialVersionUID = -1286451120913657028L;
    @Column(name = "id", length = 16, nullable = false, updatable = false)
    @GenericGenerator(name = "uuid", strategy = "uuid2")
    @GeneratedValue(generator = "uuid")
    @Basic(fetch = FetchType.EAGER)
    @Id
    protected UUID id;
    @OneToMany(mappedBy = "a", fetch = FetchType.EAGER)
    @Cascade({ CascadeType.ALL })
    List<B> bList;
    // getter + setter
}
What am I doing wrong?!
Class B:
@Entity
@Table(name = "B")
public class B implements Serializable {
    @Column(name = "id", length = 16, nullable = false, updatable = false)
    @GenericGenerator(name = "uuid", strategy = "uuid2")
    @GeneratedValue(generator = "uuid")
    @Basic(fetch = FetchType.EAGER)
    @Id
    protected UUID id;
    @ManyToOne(optional = false)
    @JoinColumns({ @JoinColumn(name = "A_id", referencedColumnName = "id", nullable = false) })
    @Valid
    A a;
    // setter + getter
}
Setters and getters are all just as simple as possbile:
public List<B> getBList() {
    return bList;
}
public void setBList(List<B> bList) {
    this.bList = bList;
}
Some more information:
Update the A.bList property as follows:
public class A {
    @OneToMany(mappedBy = "a", fetch = FetchType.EAGER, cascade = CascadeType.ALL, orphanRemoval = true)
    List<B> bList;
}
The orphanRemoval = true annotation attribute will tell the underlying JPA implementation to delete B records which don't have any parent left.
Also, Since the B side manages the association, you should clear its a attribute when breaking the relationship. To make this easier to read and to remove the burden of such implementation details from the caller, you should introduce management methods in A :
public class A {
    public void clearBList() {
        for (B b : bList) {
            b.releaseA();
        }
        bList.clear();
    }
}
public class B {
    void releaseA() {
        this.a = null;
    }
}
You should avoid exposing collections directly and instead return an immutable version of it to prevent clients to the A class of modifying the collection directly without the A class knowing it. A manages the B list, so it should have full control over it!
public class A {
    public List<B> getBList() {
        return Collections.unmodifiableList(bList);
    }
}
Hope that helps.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With