How can I get actual child collection, when adding new one in separated transactional method, while updating parent.
I have spring boot app with hibernate/jpa and one-to-many unidirectional model:
parent:
#Entity
public class Deal {
private UUID id;
#OneToMany(cascade = CascadeType.ALL, fetch = FetchType.LAZY)
private List<Rate> rates;
....
}
child:
#Entity
public class Rate {
private UUID id;
....
}
And I have non transactional method for do some business logic by rest call:
public Deal applyDeal(UUID dealId) {
dealService.apply(dealId);
return dealService.getById(dealId);
}
Method apply in DealService has several methods in separate transactions (all methods doLogic() annotated with #Transactional(Propagation.REQUIRES_NEW):
public void apply(UUI dealId) {
someService1.do1Logic(...);
someService2.do2Logic(...);
someService3.do3Logic(...);
}
In do2Logic() I have some logic that adding new Rate entity to my parent entity with dealId and direct call of save method for Deal object.
#Transactional(Propagation.REQUIRES_NEW)
publid void do2Logic(...) {
...
var deal = dealService.getById(...);
deal.getRates().add(new Rate());
dealService.save(deal);
}
But when I get response from root method applyDeal the new child entity is absent.
If after that I will try to get this parent in separate rest call (getDeal) I get actual parent entity with new child in collection.
How to get actual child collection in parent response of applyDeal method?
I tried to make all logic in one #Transactional but it doesn't works.
I also don't understand why when I am try to get deal instance to return in applyDeal I get old data.
Thank you.
I guess you are running MySQL or MariaDB? These two database by default use the repeatable read transaction isolation level, which can cause this behavior. Try configuring the read committed isolation level instead, and/or remove the REQUIRES_NEW propagation if possible, since that will suspend an already running transaction to start a second one.
I have this association in the DB -
I want the data to be persisted in the tables like this -
The corresponding JPA entities have been modeled this way (omitted getters/setters for simplicity) -
STUDENT Entity -
#Entity
#Table(name = "student")
public class Student {
#Id
#SequenceGenerator(name = "student_pk_generator", sequenceName =
"student_pk_sequence", allocationSize = 1)
#GeneratedValue(strategy = GenerationType.SEQUENCE, generator =
"student_pk_generator")
#Column(name = "student_id", nullable = false)
private Long studentId;
#Column(name = "name", nullable = false)
private String studentName;
#OneToMany(mappedBy = "student", cascade = CascadeType.ALL)
private Set<StudentSubscription> studentSubscription;
}
STUDENT_SUBSCRIPTION Entity -
#Entity
#Table(name = "student_subscription")
#Inheritance(strategy = InheritanceType.JOINED)
public abstract class StudentSubscription {
#Id
private Long studentId;
#ManyToOne(optional = false)
#JoinColumn(name = "student_id", referencedColumnName = "student_id")
#MapsId
private Student student;
#Column(name = "valid_from")
private Date validFrom;
#Column(name = "valid_to")
private Date validTo;
}
LIBRARY_SUBSCRIPTION Entity -
#Entity
#Table(name = "library_subscription",
uniqueConstraints = {#UniqueConstraint(columnNames = {"library_code"})})
#PrimaryKeyJoinColumn(name = "student_id")
public class LibrarySubscription extends StudentSubscription {
#Column(name = "library_code", nullable = false)
private String libraryCode;
#PrePersist
private void generateLibraryCode() {
this.libraryCode = // some logic to generate unique libraryCode
}
}
COURSE_SUBSCRIPTION Entity -
#Entity
#Table(name = "course_subscription",
uniqueConstraints = {#UniqueConstraint(columnNames = {"course_code"})})
#PrimaryKeyJoinColumn(name = "student_id")
public class CourseSubscription extends StudentSubscription {
#Column(name = "course_code", nullable = false)
private String courseCode;
#PrePersist
private void generateCourseCode() {
this.courseCode = // some logic to generate unique courseCode
}
}
Now, there is a Student entity already persisted with the id let's say - 100.
Now I want to persist this student's library subscription. For this I have created a simple test using Spring DATA JPA repositories -
#Test
public void testLibrarySubscriptionPersist() {
Student student = studentRepository.findById(100L).get();
StudentSubscription librarySubscription = new LibrarySubscription();
librarySubscription.setValidFrom(//some date);
librarySubscription.setValidTo(//some date);
librarySubscription.setStudent(student);
studentSubscriptionRepository.save(librarySubscription);
}
On running this test I am getting the exception -
org.springframework.dao.InvalidDataAccessApiUsageException: detached entity passed to persist: com.springboot.data.jpa.entity.Student; nested exception is org.hibernate.PersistentObjectException: detached entity passed to persist: com.springboot.data.jpa.entity.Student
To fix this I attach a #Transactional to the test. This fixed the above exception for detached entity, but the entity StudentSubscription and LibrarySubscription are not getting persisted to the DB. In fact the transaction is getting rolled back.
Getting this exception in the logs -
INFO 3515 --- [ main] o.s.t.c.transaction.TransactionContext : Rolled back transaction for test: [DefaultTestContext#35390ee3 testClass = SpringDataJpaApplicationTests, testInstance = com.springboot.data.jpa.SpringDataJpaApplicationTests#48a12036, testMethod = testLibrarySubscriptionPersist#SpringDataJpaApplicationTests, testException = [null], mergedContextConfiguration = [MergedContextConfiguration#5e01a982 testClass = SpringDataJpaApplicationTests, locations = '{}', classes = '{class com.springboot.data.jpa.SpringDataJpaApplication}', contextInitializerClasses = '[]', activeProfiles = '{}', propertySourceLocations = '{}', propertySourceProperties = '{org.springframework.boot.test.context.SpringBootTestContextBootstrapper=true}', contextCustomizers = set[org.springframework.boot.test.context.filter.ExcludeFilterContextCustomizer#18ece7f4, org.springframework.boot.test.json.DuplicateJsonObjectContextCustomizerFactory$DuplicateJsonObjectContextCustomizer#264f218, org.springframework.boot.test.mock.mockito.MockitoContextCustomizer#0, org.springframework.boot.test.web.client.TestRestTemplateContextCustomizer#2462cb01, org.springframework.boot.test.autoconfigure.actuate.metrics.MetricsExportContextCustomizerFactory$DisableMetricExportContextCustomizer#928763c, org.springframework.boot.test.autoconfigure.properties.PropertyMappingContextCustomizer#0, org.springframework.boot.test.autoconfigure.web.servlet.WebDriverContextCustomizerFactory$Customizer#7c3fdb62, org.springframework.boot.test.context.SpringBootTestArgs#1, org.springframework.boot.test.context.SpringBootTestWebEnvironment#1ad282e0], contextLoader = 'org.springframework.boot.test.context.SpringBootContextLoader', parent = [null]], attributes = map['org.springframework.test.context.event.ApplicationEventsTestExecutionListener.recordApplicationEvents' -> false]]
Now I have couple of questions -
Why am I getting detached entity exception. When we fetch an entity from the DB, Spring Data JPA must be using entityManager to fetch the entity. The fetched entity gets automatically attached to the persistence context right ?
On attaching #Transactional on the test, why the transaction is getting rolledback, and no entity is getting persisted. I was expecting the two entities - StudentSubscription and LibrarySubscription should've been persisted using the joined table inheritance approach.
I tried many things but no luck. Seeking help from, JPA and Spring DATA experts :-)
Thanks in advance.
Let me add a few details that outline a couple of design problems with your code that significantly complicate the picture. In general, when working with Spring Data, you cannot simply look at your tables, create cookie-cutter entities and repositories for those and expect things to simply work. You need to at least spend a bit of time to understand the Domain-Driven Design building blocks entity, aggregate and repository.
Repositories manage aggregates
In your case, Student treats StudentSubscriptions like an entity (full object reference, cascading persistence operations) but at the same time a repository to persist the …Subscriptions exists. This fundamentally breaks the responsibility of keeping consistency of the Student aggregate, as you can simply remove a …Subscription from the store via the repository without the aggregate having a chance to intervene. Assuming the …Subscriptions are aggregates themselves, and you'd like to keep the dependency in that direction, those must only be referred to via identifiers, not via full object representations.
The arrangement also adds cognitive load, as there are now two ways to add a subscription:
Create a …Subscription instance, assign the Student, persist the subscription via the repository.
Load a Student, create a …Subscription, add that to the student, persist the Student via it's repository.
While that's already a smell, the bidirectional relationship between the …Subscription and Student imposes the need to manually manage those in code. Also, the relationships establish a dependency cycle between the concepts, which makes the entire arrangement hard to change. You already see that you have accumulated a lot of (mapping) complexity for a rather simple example.
What would better alternatives look like?
Option 1 (less likely): Students and …Subscriptions are "one"
If you'd like to keep the concepts close together and there's no need to query the subscriptions on their own, you could just avoid those being aggregates and remove the repository for them. That would allow you to remove the back-reference from …Subscription to Student and leave you with only one way of adding subscriptions: load the Student, add a …Subscription instance, save the Student, done. This also gives the Student aggregate its core responsibility back: enforcing invariants on its state (the set of …Subscription having to follow some rules, e.g. at least one selected etc.)
Option 2 (more likely): Students and …Subscriptions are separate aggregates (potentially from separate logical modules)
In this case, I'd remove the …Subscriptions from the Student entirely. If you need to find a Students …Subscriptions, you can add a query to the …SubscriptionRepository (e.g. List<…Subscription> findByStudentId(…)). As a side effect of this you remove the cycle and Student does not (have to) know anything about …Subscriptions anymore, which simplifies the mapping. No wrestling with eager/lazy loading etc. In case any cross-aggregate rules apply, those would be applied in an application service fronting the SubscriptionRepository.
Heuristics summarized
Clear distinction between what's an aggregate and what not (the former get a corresponding repository, the later don't)
Only refer to aggregates via their identifiers.
Avoid bidirectional relationships. Usually, one side of the relationship can be replaced with a query method on a repository.
Try to model dependencies from higher-level concepts to lower level ones (Students with Subscriptionss probably make sense, a …Subscription without a Student most likely doesn't. Thus, the latter is the better relationship to model and solely work with.)
The transaction is getting rolled back because the test is doing DB updates in the test method.
#Transactional does auto rollback if the transaction includes any update DB. Also here is the compulsion to use transaction because EntityManager gets closed as soon as the Student entity gets retrieved, so to keep that open the test has to be within the transactional context.
Probably if I had used a testDB for my testcases then probably spring wouldn't haveve been rolling back this update.
Will setup an H2 testDb and perform the same operation there and will post the outcome.
Thanks for the quick help guys. :-)
Why am I getting detached entity exception. When we fetch an entity from the DB, Spring Data JPA must be using entityManager to fetch the entity. The fetched entity gets automatically attached to the persistent context right ?
Right, but only for as long as the entityManager stays open. Without the transactional, as soon as you return from studentRepository.findById(100L).get();, the entityManager gets closed and the object becomes detached.
When you call the save, a new entityManager gets created that doesn't contain a reference to the previous object. And so you have the error.
The #Trannsaction makes the entity manager stay open for the duration of the method.
At least, that's what I think it's happening.
On attaching #Transactional on the test, why the transaction is getting rolledback,
With bi-directional associations, you need to make sure that the association is updated on both sides. The code should look like:
#Test
#Transactional
public void testLibrarySubscriptionPersist() {
Student student = studentRepository.findById(100L).get();
StudentSubscription librarySubscription = new LibrarySubscription();
librarySubscription.setValidFrom(//some date);
librarySubscription.setValidTo(//some date);
// Update both sides:
librarySubscription.setStudent(student);
student.getStudentSubscription().add(librarySubscription);
// Because of the cascade, saving student should also save librarySubscription.
// Maybe it's not necessary because student is managed
// and the db will be updated anyway at the end
// of the transaction.
studentSubscriptionRepository.save(student);
}
In this case, you could also use EntityManager#getReference:
#Test
#Transactional
public void testLibrarySubscriptionPersist() {
EntityManager em = ...
StudentSubscription librarySubscription = new LibrarySubscription();
librarySubscription.setValidFrom(//some date);
librarySubscription.setValidTo(//some date);
// Doesn't actually load the student
Student student = em.getReference(Student.class, 100L);
librarySubscription.setStudent(student);
studentSubscriptionRepository.save(librarySubscription);
}
I think any of these solutions should fix the issue. Hard to say without the whole stacktrace.
I'm working with Spring Boot 1.3.0.M4 and a MySQL database.
I have a problem when using modifying queries, the EntityManager contains outdated entities after the query has executed.
Original JPA Repository:
public interface EmailRepository extends JpaRepository<Email, Long> {
#Transactional
#Modifying
#Query("update Email e set e.active = false where e.active = true and e.expire <= NOW()")
Integer deactivateByExpired();
}
Suppose we have Email [id=1, active=true, expire=2015/01/01] in DB.
After executing:
emailRepository.save(email);
emailRepository.deactivateByExpired();
System.out.println(emailRepository.findOne(1L).isActive()); // prints true!! it should print false
First approach to solve the problem: add clearAutomatically = true
public interface EmailRepository extends JpaRepository<Email, Long> {
#Transactional
#Modifying(clearAutomatically = true)
#Query("update Email e set e.active = false where e.active = true and e.expire <= NOW()")
Integer deactivateByExpired();
}
This approach clears the persistence context not to have outdated values, but it drops all non-flushed changes still pending in the EntityManager. As I use only save() methods and not saveAndFlush() some changes are lost for other entities :(
Second approach to solve the problem: custom implementation for repository
public interface EmailRepository extends JpaRepository<Email, Long>, EmailRepositoryCustom {
}
public interface EmailRepositoryCustom {
Integer deactivateByExpired();
}
public class EmailRepositoryImpl implements EmailRepositoryCustom {
#PersistenceContext
private EntityManager entityManager;
#Transactional
#Override
public Integer deactivateByExpired() {
String hsql = "update Email e set e.active = false where e.active = true and e.expire <= NOW()";
Query query = entityManager.createQuery(hsql);
entityManager.flush();
Integer result = query.executeUpdate();
entityManager.clear();
return result;
}
}
This approach works similar to #Modifying(clearAutomatically = true) but it first forces the EntityManager to flush all changes to DB before executing the update and then it clears the persistence context. This way there won't be outdated entities and all changes will be saved in DB.
I would like to know if there's a better way to execute update statements in JPA without having the issue of the outdated entities and without the manual flush to DB. Perhaps disabling the 2nd level cache? How can I do it in Spring Boot?
Update 2018
Spring Data JPA approved my PR, there's a flushAutomatically option in #Modifying() now.
#Modifying(flushAutomatically = true, clearAutomatically = true)
I know this is not a direct answer to your question, since you already have built a fix and started a pull request on Github. Thank you for that!
But I would like to explain the JPA way you can go. So you would like to change all entities which match a specific criteria and update a value on each. The normal approach is just to load all needed entities:
#Query("SELECT * FROM Email e where e.active = true and e.expire <= NOW()")
List<Email> findExpired();
Then iterate over them and update the values:
for (Email email : findExpired()) {
email.setActive(false);
}
Now hibernate knows all changes and will write them to the database if the transaction is done or you call EntityManager.flush() manually. I know this won't work well if you have a big amount of data entries, since you load all entities into memory. But this is the best way, to keep the hibernate entity cache, 2nd level caches and the database in sync.
Does this answer say "the `#Modifying´ annotation is useless"? No! If you ensure the modified entities are not in your local cache e.g. write-only application, this approach is just the way to go.
And just for the record: you don't need #Transactional on your repository methods.
Just for the record v2: the active column looks as it has a direct dependency to expire. So why not delete active completely and look just on expire in every query?
As klaus-groenbaek said, you can inject EntityManager and use its refresh method :
#Inject
EntityManager entityManager;
...
emailRepository.save(email);
emailRepository.deactivateByExpired();
Email email2 = emailRepository.findOne(1L);
entityManager.refresh(email2);
System.out.println(email2.isActive()); // prints false
In one of our spring batch jobs, we create additional entities (CompanyProfile) during processing and persist them to the DB (in a separate transaction). These entities are referenced by other entities (Vacancy), which will be persisted by the writer, but unfortunate the writer fails with this error:
Caused by: javax.persistence.EntityNotFoundException: Unable to find com.company.CompanyProfile with id 1409881
The model is as follows:
#Entity
public class Vacancy {
#ManyToOne(fetch = FetchType.EAGER, optional = true)
#JoinColumn(name = "company", nullable = true)
private CompanyProfile company;
...
}
#Entity
public class CompanyProfile {
#Id
#GeneratedValue(strategy = GenerationType.AUTO)
private Long id;
...
}
In the processor we have this:
CompanyProfile company = companyProfileService.handleCompany(compName);
vacancy.setCompany(company);
Where the method companyProfileService.handleCompany() is annotated with #Transactional(readOnly = false, propagation = Propagation.REQUIRES_NEW )
I'm sure the CompanyProfile gets persisted - I can see it in the DB, but when the Vacancy gets saved by the ItemWriter, it fails with the above exception. (also, note that the id of the persisted entity is mention in the exception above)
Do you see any reason why the writer would fail in this case?
With information you gave us my guess is that transaction opened by SB is unable to see data persisted by companyProfileService.handleCompany() method because service component uses a different transaction than SB ones; you have to check database ISOLATION_LEVEL property
TL;DR: Is it possible to perform nested transactions in Hibernate that use SavePoints to rollback to specific states?
So I am attempting to persist a parent entity with a OneToMany mapping to child entities. This is working fine.
During this persistence, I would like to catch and log ALL constraint violations that occur. Currently, the FIRST entity (child or parent) to have a constraint violation throws a ConstraintViolationException and rolls back the transaction. I would like for the transaction to still be rolled back, but somehow collect ALL of the constraint violations that would occur.
Here is a brief outline of my entities:
ParentEntity.java
#Entity
#Table(name = "PARENT", schema = "SOMESCHEMA")
public class ParentEntity {
private static final ID_COLUMN = "ID_COLUMN";
#Id
#Column(name = ID_COLUMN)
private Long id;
#OneToMany(fetch = FetchType.LAZY, cascade = {CascadeType.ALL}, orphanRemoval = true)
#JoinColumn(name = ID_COLUMN, referencedColumnName = ID_COLUMN)
private List<childEntity> children;
}
ChildEntity.java
#Entity
#Table(name = "CHILD", schema = "SOMESCHEMA")
public class ChildEntity {
public ChildEntity(String input) {
this.validationString = input;
}
#Id
#Column(name = ParentEntity.ID_COLUMN)
private Long id;
#ManyToOne
#JoinColumn(name = ParentEntity.ID_COLUMN, insertable = false, updatable = false)
private ParentEntity parent;
// The field under validation (should be less than 25 char's long)
#Column(name = "VALIDATE_ME")
private String validationString;
}
Example run:
public void someMethod() {
ParentEntity parent = new ParentEntity();
parent.addChild(new Child("good input 1"));
parent.addChild(new Child("bad input 1 break here"));
parent.addChild(new Child("bad input 2 break here"));
parent.addChild(new Child("good input 2"));
dataAccessObject.persist(parent);
}
Results:
I see the transaction rolled back and the ConstraintViolationException only contains information for the first bad child.
Desired Results:
I see the transaction rolled back and the ConstraintViolationException show information for all the bad children regardless of how many children were bad. (Also, if the parent has a constraint violation, I would still like to check the child constraints)
Is this possible?
ConstraintViolationException is a HibernateException and all Hibernate exceptions are not recoverable. So if one exception is thrown, you cannot rely on the existing session state to continue processing reliably.
So this is not possible or recommended.
I found a way to accomplish my end result, although it was through means I did not expect.
In order to accomplish my goal, I would need to use nested transactions and SavePoints. At the end of the day, that implementation would still produce invalid data in my database, because during the time it takes to find an error in a child, some of the other children may have already been persisted and a consumer of my database would be unaware that the parent and all it's children were about to be deleted due to one or more bad entities (parents or children).
My solution:
I implemented a validator to validate all of the parents and children before going to persist them. The drawback to this method is that I have to annotate constraints on my entity fields, but double validation is never a bad thing.
The answer to my original question:
This is impossible to due with my version of Hibernate, unless I implemented custom SavePoint functionality to support nested transactions.