How do I update an entity using spring-data-jpa?
Asked Answered
L

17

332

Well the question pretty much says everything. Using JPARepository how do I update an entity?

JPARepository has only a save method, which does not tell me if it's create or update actually. For example, I insert a simple Object to the database User, which has three fields: firstname, lastname and age:

 @Entity
 public class User {

  private String firstname;
  private String lastname;
  //Setters and getters for age omitted, but they are the same as with firstname and lastname.
  private int age;

  @Column
  public String getFirstname() {
    return firstname;
  }
  public void setFirstname(String firstname) {
    this.firstname = firstname;
  }

  @Column
  public String getLastname() {
    return lastname;
  }
  public void setLastname(String lastname) {
    this.lastname = lastname;
  }

  private long userId;

  @Id
  @GeneratedValue(strategy=GenerationType.AUTO)
  public long getUserId(){
    return this.userId;
  }

  public void setUserId(long userId){
    this.userId = userId;
  }
}

Then I simply call save(), which at this point is actually an insert into database:

 User user1 = new User();
 user1.setFirstname("john"); user1.setLastname("dew");
 user1.setAge(16);

 userService.saveUser(user1);// This call is actually using the JPARepository: userRepository.save(user);

So far so good. Now I want to update this user, say change his age. For this purpose I could use a Query, either QueryDSL or NamedQuery, whatever. But, considering I just want to use spring-data-jpa and the JPARepository, how do I tell it that instead of an insert I want to do an update?

Specifically, how do I tell spring-data-jpa that users with the same username and firstname are actually EQUAL and that the existing entity supposed to be updated? Overriding equals did not solve this problem.

Leucotomy answered 9/8, 2012 at 10:33 Comment(5)
Are you sure the ID gets rewritten when you save an existing object in the database?? Never had this on my project tbhPiscatorial
@ByronVoorbach yup you're right, just tested this. update the question also, thxLeucotomy
Hello friend, you can look this link #24421072 you can be a approach like saveOrUpdate()Potheen
#39741602Muley
I think that we have a beautiful solution here: enter link description hereArmoire
M
300

Identity of entities is defined by their primary keys. Since firstname and lastname are not parts of the primary key, you cannot tell JPA to treat Users with the same firstnames and lastnames as equal if they have different userIds.

So, if you want to update a User identified by its firstname and lastname, you need to find that User by a query, and then change appropriate fields of the object your found. These changes will be flushed to the database automatically at the end of transaction, so that you don't need to do anything to save these changes explicitly.

EDIT:

Perhaps I should elaborate on overall semantics of JPA. There are two main approaches to design of persistence APIs:

  • insert/update approach. When you need to modify the database you should call methods of persistence API explicitly: you call insert to insert an object, or update to save new state of the object to the database.

  • Unit of Work approach. In this case you have a set of objects managed by persistence library. All changes you make to these objects will be flushed to the database automatically at the end of Unit of Work (i.e. at the end of the current transaction in typical case). When you need to insert new record to the database, you make the corresponding object managed. Managed objects are identified by their primary keys, so that if you make an object with predefined primary key managed, it will be associated with the database record of the same id, and state of this object will be propagated to that record automatically.

JPA follows the latter approach. save() in Spring Data JPA is backed by merge() in plain JPA, therefore it makes your entity managed as described above. It means that calling save() on an object with predefined id will update the corresponding database record rather than insert a new one, and also explains why save() is not called create().

Munoz answered 9/8, 2012 at 10:43 Comment(12)
well yes I think I know this. I was strictly referring to spring-data-jpa. I have two problems with this answer now: 1) business values are not supposed to be part of the primary key - that is a known thing, right? So having the firstname and lastname as primary key is not good. And 2) Why is this method then not called create, but save instead in spring-data-jpa?Leucotomy
"save() in Spring Data JPA is backed by merge() in plain JPA" did you actually look at the code? I just did and it both backed up by either persist or merge. It will persist or update based on the presence of the id (primary key). This, I think, should be documented in the save method. So save is actually EITHER merge or persist.Leucotomy
I think it is also called save, because it it supposed to save the object no matter what state it is in - it is going to perform an update or insert which is equal to save state.Leucotomy
@Eugene: Yes, I did. As far as I understand it doesn't matter, because persist() is called when entity is new, i.e. when merge() works as persist().Munoz
yup, I guess we are talking about the same thing, in different words. Thx for your time, points given.Leucotomy
This will not work for me. I have tried save on an entity with a valid primary key. I access the page with "order/edit/:id" and it actually gives me the correct object by Id. Nothing I try for the love of God will update the entity. It always posts a new entity .. I even tried making a custom service and using "merge" with my EntityManager and it still won't work. It will always post a new entity.Hassiehassin
@DTechNet I was experiencing a similar problem to you, DtechNet, and it turned out my problem was that I had the wrong primary key type specified in my Spring Data repository interface. It said extends CrudRepository<MyEntity, Integer> instead of extends CrudRepository<MyEntity, String> like it should have. Does that help? I know this is almost a year later. Hope it helps someone else.Twine
@KentJohnson why do you store the primary key as a String? I thought you only ever used Long or Integer.Hassiehassin
@DTechNet it made business sense to those who made the decision. It is a natural key for the dataset so those who made the decision thought it was a good idea. It is for uniquely identifying an object that has a String of digits as its physical identifier.Twine
Alright. I don't plan on using a String though. I plan on using Integer. Perhaps Long.Hassiehassin
Your answer is very inspiring thanks. if Spring data jpa save() is actually merge() in JPA, why bother CascadeType.Persist on @XtoX relationship?Boxthorn
Curious question, but when is said transaction end? Does it end when the calling method finishes?Subtend
C
238

Since the answer by @axtavt focuses on JPA not spring-data-jpa

To update an entity by querying then saving is not efficient because it requires two queries and possibly the query can be quite expensive since it may join other tables and load any collections that have fetchType=FetchType.EAGER

Spring-data-jpa supports update operation.
You have to define the method in Repository interface.and annotated it with @Query and @Modifying.

@Modifying
@Query("update User u set u.firstname = ?1, u.lastname = ?2 where u.id = ?3")
void setUserInfoById(String firstname, String lastname, Integer userId);

@Query is for defining custom query and @Modifying is for telling spring-data-jpa that this query is an update operation and it requires executeUpdate() not executeQuery().

You can specify the return type as int, having the number of records being updated.


Note: Run this code in a Transaction.

Carousal answered 30/6, 2015 at 22:28 Comment(17)
Make sure you run it in transactionCarousal
Hey! Thanks am using spring data beans. So it will automatically takes care of my update. <S extends T> S save(S entity); automatically takes care of update. i didnt have to use your method! Thanks anyways!Edp
Anytime :) The save method does work if you want to save the entity (It will delegate the call to either em.persist() or em.merge() behind the scene). Anyway, the custom query is useful when you want to update just some fields in database.Carousal
how about when one of your parameters is id of sub entity( manyToOne) how should update that? (book have an Author and you passed book id and author id to update book author)Kellikellia
To update an entity by querying then saving is not efficient these are not the only two choices. There is a way to specify id and get the row object without querying it. If you do a row = repo.getOne(id) and then row.attr = 42; repo.save(row); and watch the logs, you will see only the update query.Kile
@Kile Thank you for your comment. I'm just curious. Is that vendor specific optimization? I mean will it work across different JPA imps? and what if you perform a complex query instead of querying by ID, will it still generates a single update query?Carousal
@Carousal you can always generate a reference to a row using it's ID. For more complex queries like inner join updates or updates with dissimilar where conditions, it won't be useful.Kile
First, It's not a good practice to eagerly load a collection. Second, if we choose to use Modifiying, we should also manually update the version. So, I don't think it's a good option to update a single entity instance using Modifying if performance is not a real matter. Modifying should be used only for batch updates.Fissirostral
@Carousal - Is there any way if we can count=1 of update if successful? Then its very best to implementQuenby
Thanks just one question - How will I know that the specific entry is updated or not? Can we return something to know it is successfully updated or not found or given input value?Host
Updated the answer. You can return int or boolean as well.Carousal
@Transactional add on top of the query.Pandemonium
Will this still be beneficial if we need to do multiple updates at once as opposed to using saveAll()?Sundog
Modifying queries can only use void or int/Integer as return type. boolean will not work hereGrotto
How can i make spring jpa throw exception when the where clause in update query is not specified. ie the number of updated row is zero?Averill
How do you determine whether to insert a new item or update though? I'm assuming this example is only beneficial when you know you will be updating an existing row..Hjerpe
the documentation for <a href="docs.spring.io/spring-data/jpa/docs/current-SNAPSHOT/reference/… @Query</a> and for <a href="docs.spring.io/spring-data/jpa/docs/current-SNAPSHOT/reference/…>Sevilla
E
65

You can simply use this function with save() JPAfunction, but the object sent as parameter must contain an existing id in the database otherwise it will not work, because save() when we send an object without id, it adds directly a row in database, but if we send an object with an existing id, it changes the columns already found in the database.

public void updateUser(Userinfos u) {
    User userFromDb = userRepository.findById(u.getid());
    // crush the variables of the object found
    userFromDb.setFirstname("john"); 
    userFromDb.setLastname("dew");
    userFromDb.setAge(16);
    userRepository.save(userFromDb);
}
Effeminate answered 30/9, 2018 at 10:20 Comment(6)
isn't the performance a problem if you have to load the object from database before do the update? (sorry for my english)Surratt
there is two query instead of one, which is highly not prefferedAlpha
i know i juste showed up another method to do ! but why the jpa implemented the update function when the id is the same ?Effeminate
You can use userRepository.getById(u.getid()) to avoid two queries (getById instead of findById).Stannite
userRepository.getById() is deprecated.Ani
do we need to call save method? Since userFromDb object is persisten, it must be updated automatically after properties are set.Ise
L
49

As what has already mentioned by others, the save() itself contains both create and update operation.

I just want to add supplement about what behind the save() method.

Firstly, let's see the extend/implement hierarchy of the CrudRepository<T,ID>, enter image description here

Ok, let's check the save() implementation at SimpleJpaRepository<T, ID>,

@Transactional
public <S extends T> S save(S entity) {

    if (entityInformation.isNew(entity)) {
        em.persist(entity);
        return entity;
    } else {
        return em.merge(entity);
    }
}

As you can see, it will check whether the ID is existed or not firstly, if the entity is already there, only update will happen by merge(entity) method and if else, a new record is inserted by persist(entity) method.

Leucotomy answered 11/12, 2018 at 3:38 Comment(3)
should not merge be used with detached entities. And to confuse further if the entity is transient, it copies upon a newly created persistent entity. . Why use persist at all. If merge can handle all four states.Headsail
good reference to code +1Headsail
Most developers break their head creating routines to process updates when in reality all the hard work is already done. This is the best, most practical answer.Macroclimate
V
13

spring data save() method will help you to perform both: adding new item and updating an existed item.

Just call the save() and enjoy the life :))

Valle answered 3/6, 2019 at 10:6 Comment(6)
Up in this way if i sent different Id will save it , how can i avoid save new record .Stoddard
@AbdAbughazaleh check whether incoming Id exists in your repository or not. you can use ' repository.findById(id).map( entity -> { //do something return repository.save(entity) } ).orElseGet( () -> { //do something return; }); 'Valle
Using save() we can avoid adding new record in the targeted table (update operation) if that table has some unique column (say Name). Now the passed entity to be updated should have same Name value as table otherwise it will create a new record.Brochu
@AbdAbughazaleh What about #OneToMany deleting the child entities that are not available in incoming request? I'm stuck since 15 days. Can you please provide your precious inputs?Heathen
@Ram Isn't Id field sufficient? That will be the unique column right?Strode
If you call save without having modified the data it won't actually be updated.Dignify
V
10

With java 8 you can use repository's findById in UserService

@Service
public class UserServiceImpl {

    private final UserRepository repository;

    public UserServiceImpl(UserRepository repository) {
        this.repository = repository;
    }

    @Transactional
    public void update(User user) {
        repository
                .findById(user.getId()) // returns Optional<User>
                .ifPresent(user1 -> {
                    user1.setFirstname(user.getFirstname);
                    user1.setLastname(user.getLastname);

                    repository.save(user1);
                });
    }

}
Veery answered 21/6, 2022 at 20:49 Comment(0)
P
9

Using spring-data-jpa save(), I was having same problem as @DtechNet. I mean every save() was creating new object instead of update. To solve this I had to add version field to entity and related table.

Pepperandsalt answered 6/12, 2015 at 18:54 Comment(4)
what you mean by to add version field to entity and related table.Telethermometer
https://mcmap.net/q/100608/-java-jpa-version-annotationPepperandsalt
for me save() still creates a new entity even after adding version.Contagious
@ParagKadam, make sure your entity has ID on it before you saving it.Pepperandsalt
R
8

This is how I solved the problem:

User inbound = ...
User existing = userRepository.findByFirstname(inbound.getFirstname());
if(existing != null) inbound.setId(existing.getId());
userRepository.save(inbound);
Restive answered 21/9, 2017 at 15:40 Comment(1)
Use @Transaction above method for several db request. An in this case no need in userRepository.save(inbound);, changes flushed automatically.Shevlo
A
2
public void updateLaserDataByHumanId(String replacement, String humanId) {
    List<LaserData> laserDataByHumanId = laserDataRepository.findByHumanId(humanId);
    laserDataByHumanId.stream()
            .map(en -> en.setHumanId(replacement))
            .collect(Collectors.toList())
            .forEach(en -> laserDataRepository.save(en));
}
Audre answered 26/3, 2019 at 10:42 Comment(1)
This statement needs a bit change. .map(en -> {en.setHumanId(replacement); return en;})Toratorah
E
1

Specifically how do I tell spring-data-jpa that users that have the same username and firstname are actually EQUAL and that it is supposed to update the entity. Overriding equals did not work.

For this particular purpose one can introduce a composite key like this:

CREATE TABLE IF NOT EXISTS `test`.`user` (
  `username` VARCHAR(45) NOT NULL,
  `firstname` VARCHAR(45) NOT NULL,
  `description` VARCHAR(45) NOT NULL,
  PRIMARY KEY (`username`, `firstname`))

Mapping:

@Embeddable
public class UserKey implements Serializable {
    protected String username;
    protected String firstname;

    public UserKey() {}

    public UserKey(String username, String firstname) {
        this.username = username;
        this.firstname = firstname;
    }
    // equals, hashCode
}

Here is how to use it:

@Entity
public class UserEntity implements Serializable {
    @EmbeddedId
    private UserKey primaryKey;

    private String description;

    //...
}

JpaRepository would look like this:

public interface UserEntityRepository extends JpaRepository<UserEntity, UserKey>

Then, you could use the following idiom: accept DTO with user info, extract name and firstname and create UserKey, then create a UserEntity with this composite key and then invoke Spring Data save() which should sort everything out for you.

Emoryemote answered 25/7, 2019 at 12:3 Comment(0)
P
1

You can see the example below:

private void updateDeliveryStatusOfEvent(Integer eventId, int deliveryStatus) {
    try {
        LOGGER.info("NOTIFICATION_EVENT updating with event id:{}", eventId);
        Optional<Event> eventOptional = eventRepository.findById(eventId);
        if (!eventOptional.isPresent()) {
            LOGGER.info("Didn't find any updatable notification event with this eventId:{}", eventId);
        }
        Event event = eventOptional.get();
        event.setDeliveryStatus(deliveryStatus);
        event = eventRepository.save(event);
        if (!Objects.isNull(event)) {
            LOGGER.info("NOTIFICATION_EVENT Successfully Updated with this id:{}", eventId);
        }
    } catch (Exception e) {
        LOGGER.error("Error :{} while updating NOTIFICATION_EVENT of event Id:{}", e, eventId);
    }
}

Or Update Using Native Query:

public interface YourRepositoryName extends JpaRepository<Event,Integer>{
@Transactional
    @Modifying
    @Query(value="update Event u set u.deliveryStatus = :deliveryStatus where u.eventId = :eventId", nativeQuery = true)
    void setUserInfoById(@Param("deliveryStatus")String deliveryStatus, @Param("eventId")Integer eventId);
}
Philip answered 28/12, 2021 at 10:35 Comment(0)
B
1

As mentioned by others answer, method save() is dual function. It can both do save or update, it's automatically update if you provide the id.

for update method in controller class I suggested to use @PatchMapping. below is the example.

#Save method POST

{
    "username": "jhon.doe",
    "displayName": "Jhon",
    "password": "xxxyyyzzz",
    "email": "[email protected]"
}
@PostMapping("/user")
public void setUser(@RequestBody User user) {
    userService.save(user);
}

#Update method PATCH

{
    "id": 1, // this is important. Widly important
    "username": "jhon.doe",
    "displayName": "Jhon",
    "password": "xxxyyyzzz",
    "email": "[email protected]"
}

@PatchMapping("/user")
public void patchUser(@RequestBody User user) {
    userService.save(user);
}

Maybe you're wondering where the id's come from. It comes from the database of course, you want to update the existing data right?

Birkett answered 21/4, 2022 at 2:59 Comment(2)
What is UserService business logic here? repository.save() will add new entry right.Empathic
@Thamaraiselvam, you are right, it is for saving. May be you are wondering why I'm using UserService instead of UserRepository, eventhough both class have the same function. It's for internal convention and may be "culture" of the previous experience it's always using service when handling business logic. It's not simple saving though, we are adding AOP before and after the "service" classes, like "update by", "created by", "modified date", "created date".Birkett
O
0

If your primary key is autoincrement then, you have to set the value for the primary key. for the save(); method to work as a update().else it will create a new record in db.

if you are using jsp form then use hidden filed to set primary key.

Jsp:

<form:input type="hidden" path="id" value="${user.id}"/>

Java:

@PostMapping("/update")
public String updateUser(@ModelAttribute User user) {
    repo.save(user);
    return "redirect:userlist";
}

also look at this:

@Override
  @Transactional
  public Customer save(Customer customer) {

    // Is new?
    if (customer.getId() == null) {
      em.persist(customer);
      return customer;
    } else {
      return em.merge(customer);
    }
  }
Oly answered 18/10, 2020 at 9:37 Comment(0)
L
0

Use @DynamicUpdate annotation. it is cleaner and you don't have to deal with querying the database in order to get the saved values.

Lordly answered 29/3, 2021 at 20:56 Comment(2)
Would be much helpful if you add a simple example here.Chammy
LinkFemale
E
0

just use

userRepository.save(user)

if any user of this user's primary-key (i.e. id in your case) is found then this statement will update existing user otherwise this statement will save new user.

Ednaedny answered 15/6, 2023 at 9:20 Comment(0)
E
0

Since my Transaction entity has a lot of fields this is the approach that works best for me:

  @Override
  @Transactional
  public Transaction updateTransaction(UUID id, Transaction updatedTransaction) {
    Transaction transactionToUpdate = transactionRepository.getReferenceById(id);
    BeanUtils.copyProperties(updatedTransaction, transactionToUpdate, "transactionId");
    return transactionRepository.save(transactionToUpdate);
  }
Executioner answered 14/11, 2023 at 9:51 Comment(0)
I
-1

I did this for my Entity UserModel:

In the Controller:

        @PutMapping("/{id}")
        public Optional<UserModel> update(@RequestBody UserModel user, @PathVariable Long id) {
            return this.userService.update(user, id);
        }

And in the Service:

    public Optional<UserModel> update(UserModel req, Long id){
        Optional<UserModel> user = userRepository.findById(id);
        if (user != null) {
            userRepository.save(req);
        }
        return user;
    }

Example with postman: Postman method PUT example

Io answered 16/2, 2023 at 0:21 Comment(0)

© 2022 - 2024 — McMap. All rights reserved.