Spring Data JPA query return repeated row instead

问题:

Entity class

public class Event {

    @Id
    private String name;

    private String description;

    private Date eventDateTime;

    //getter and setter code
}

Service Class

EventService {

  @Autowired EventRepository eventRepository;

  List<Event> getEvents () {
     List<Event> events = eventRepository.findAll();

     return events;
  }
}

For sample data set: Event ('add', '', '2018-01-01 00:00:10') Event ('add', '', '2018-01-01 00:10:10') Event ('delete', '', '2018-01-01 00:20:00') Event ('edit', '', '2018-01-01 00:30:00')

JPA findAll() query return repeated rows:

Event ('add', '', '2018-01-01 00:00:10') Event ('add', '', '2018-01-01 00:00:10') Event ('add', '', '2018-01-01 00:00:10') Event ('add', '', '2018-01-01 00:00:10')

回答1:

To avoid repeated (duplicate) data, we have to ensure there is a unique key and that will be annotated by @Id. In this example, name it self is not unique, that's why the result show duplicate data. eventDateTime is better choice as unique field.

public class Event {

    private String name;

    private String description;

    @Id
    private Date eventDateTime;

    //getter and setter code
}

Or, we can define a composite unique key with name and eventDateTime.

public class CompositeKey implements Serializable { 

    private String name;

    private Date eventDateTime;
}

Then, annotated Event class with @IdClass(CopositeKey.class) and both name and eventDateTime field with @Id

 @IdClass(CopositeKey.class)
 public class Event {
        @Id
        private String name;

        private String description;

        @Id
        private Date eventDateTime;

        //getter and setter code
    }

回答2:

Your name-column is the identifier of the entity (@Id), yet your sample-data contains the String data two times.

Identifiers have to be unique, if you manually assign them (i.e. not using generated identifier) it's the appliations responsibility to keep them unique.

If not strange behaviour may occur.

So you should either fix your sample data or use another column (with generated values) as the identifier of the entitiy.