I have a one to many relationship: Order and Items. An Order may have lots of items
The code below (all code) is auto generated by Spring Template project:
On the Order Side:
#OneToMany(cascade=CascadeType.ALL)
#JoinColumn(name="ORDER_ID")
private Collection<Item> items = new LinkedHashSet<Item>();
On the Item Side:
#ManyToOne
private Order order;
Test code that works:
order.getItems().add(new Item());
session.save(order)
Order other = (Order) session.get(Order.class, order.getId());
assertEquals(1, other.getItems().size());
While searching through the internet and various sites with tutorials on 1 <---> Many relationship I usually find code such as example: hibernate one to many
The One Side (Stock)
#OneToMany(fetch = FetchType.LAZY, mappedBy = "stock")
public Set<StockDailyRecord> getStockDailyRecords() {
return this.stockDailyRecords;}
The many Side (Stock Record)
#ManyToOne(fetch = FetchType.LAZY)
#JoinColumn(name = "STOCK_ID", nullable = false)
public Stock getStock() {
return this.stock;}
Why is the JoinColumn Annotation placed differently?The owner of the relationship according to spring code is on the side of the One and not the many. why is that?
Althouth the relationship does not have the mappedBy atttibute at all so one could say that is one direction from order to items it seems bidirectional. i.e order.getItems().add(new Item) and order.getItems().iterator().next().getOrder() works as well!! On top of that the tables created in the actual database seem ok.
If we refactor the first code (see below the refactored) and place the annotations as the tutorials suggest for bidirectional mapping then 2/3 of the tests fail because no items are saved in the database (Assertion error expected 1 but was 0 when using order.getItems().add(new Item())) .Note that the spring tempate test code is left unchanged.Why?
Does it have to do with the annotations being placed on get methods vs annotations placed on private fields?
What is the proper way of saving Items and Order in database (java code)?
Thanks
The code refactored that fails the tests:
The One Side:
#OneToMany(cascade=CascadeType.ALL,mappedBy="order")
private Collection<Item> items = new LinkedHashSet<Item>();
The many side:
#ManyToOne
#JoinColumn(name="ORDER_ID")
private Order order;
Related
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 have a entity class which has multiple children with oneToMany association:
public class A{
private Long id;
private String name;
#OneToMany(mappedBy = "A", fetch = FetchType.LAZY, cascade = CascadeType.ALL,
orphanRemoval = true)
private List<B>bList= new ArrayList<>();
#OneToMany(mappedBy = "A", fetch = FetchType.LAZY, cascade = CascadeType.ALL,
orphanRemoval = true)
private List<C>cList= new ArrayList<>();
#OneToMany(mappedBy = "A", fetch = FetchType.LAZY, cascade = CascadeType.ALL,
orphanRemoval = true)
private List<D>dList= new ArrayList<>();
//getters and setters
}
For B,C and D I have set ManyToOne. In a word, they are in a bi-directional relationship.
Now, If I fetch A by id, I see a lot of queries get fired which turns out to be N+1 problem. To solve this, I added #Fetch(FetchMode.SUBSELECT) to all of the oneToMany relationships above which cause less queries to be fired.
My question is:
is it okay using #Fetch(FetchMode.SUBSELECT) or I can optimize it further?
what if I want to fetch all "As" by calling findAll() method? What should be the syntax for multiple children? Like
"select a from A a join fetch a.b then ??"
List< A > findAll()
Now, If I fetch A by id, I see a lot of queries get fired which turns out to be N+1 problem. To solve this, I added #Fetch(FetchMode.SUBSELECT) to all of the oneToMany relationships above which cause less queries to be fired.
You are not saying how/when these queries are fired, so a possible reason for the problem is that you are returning entities from your HTTP endpoints which are then serialized. Using #Fetch(FetchMode.SUBSELECT) is one way to "improve" the performance but will only work nicely if the base query which you use to fetch A is simple. If it gets too complex (pagination, complex predicates, etc.) you should stick to the default which is SELECT and instead configure a proper batch size via the #BatchSize( size = 32 ) annotation. A good value for the batch size would be the amount of A instances that you expect to be returned such that only a single query is executed per collection. If you allow a max page size of e.g. 50, you setting the value to 50 would be perfect.
List< A > findAll()
Do not ever do this if you care about performance and usability. It rarely makes sense to allow returning all elements, as no user can handle more than ~20 elements at a time anyway. As mentioned before, always have some kind of upper limit on the page size to prevent misuses that can cause performance issues.
I have two entities Questions and UserAnswers. I need to make an api in spring boot which returns all the columns from both the entities based on some conditions.
Conditions are:
I will be give a comparator eg: >, <, =, >=, <=
A column name eg: last_answered_at, last_seen_at
A value of the above column eg: 28-09-2020 06:00:18
I will need to return an inner join of the two entities and filter based on the above conditions.
Sample sql query based on above conditions will be like:
SELECT q,ua from questions q INNER JOIN
user_answers ua on q.id = ua.question_id
WHERE ua.last_answered_at > 28-09-2020 06:00:18
The problem I am facing is that the column name and the comparator for the query needs to be dynamic.
Is there an efficient way to do this using spring boot and JPA as I do not want to make jpa query methods for all possible combinations of columns and operators as it can be a very large number and there will be extensive use of if else?
I have developed a library called spring-dynamic-jpa to make it easier to implement dynamic queries with JPA.
You can use it to write the query templates. The query template will be built into different query strings before execution depending on your parameters when you invoke the method.
This sounds like a clear custom implementation of a repository method. Firstly, I will make some assumptions about the implementation of your entities. Afterwards, I will present an idea on how to solve your challenge.
I assume that the entities look basically like this (getters, setters, equals, hachCode... ignored).
#Entity
#Table(name = "questions")
public class Question {
#Id
#GeneratedValue
private Long id;
private LocalDateTime lastAnsweredAt;
private LocalDateTime lastSeenAt;
// other attributes you mentioned...
#OneToMany(mappedBy = "question", cascade = CascadeType.ALL, orphanRemoval = true)
private List<UserAnswer> userAnswers = new ArrayList();
// Add and remove methods added to keep bidirectional relationship synchronised
public void addUserAnswer(UserAnswer userAnswer) {
userAnswers.add(userAnswer);
userAnswer.setQuestion(this);
}
public void removeUserAnswer(UserAnswer userAnswer) {
userAnswers.remove(userAnswer);
userAnswer.setQuestion(null);
}
}
#Entity
#Table(name = "user_answers")
public class UserAnswer {
#Id
#GeneratedValue
private Long id;
#ManyToOne(fetch = FetchType.LAZY)
#JoinColumn(name = "task_release_id")
private Question question;
}
I will write the code with the knowledge about the JPA of Hibernate. For other JPAs, it might work similarly or the same.
Hibernate often needs the name of attributes as a String. To circumvent the issue of undetected mistakes (especially when refactoring), I suggest the module hibernate-jpamodelgen (see the class names suffixed with an underscore). You can also use it to pass the names of the attributes as arguments to your repository method.
Repository methods try to communicate with the database. In JPA, there are different ways of implementing database requests: JPQL as a query language and the Criteria API (easier to refactor, less error prone). As I am a fan of the Criteria API, I will use the Criteria API together with the modelgen to tell the ORM Hibernate to talk to the database to retrieve the relevant objects.
public class QuestionRepositoryCustomImpl implements QuestionRepository {
#PersistenceContext
private EntityManager entityManager;
#Override
public List<Question> dynamicFind(String comparator, String attribute, String value) {
CriteriaBuilder cb = entityManager.getCriteriaBuilder();
CriteriaQuery<Question> cq = cb.createQuery(Question.class);
// Root gets constructed for first, main class in the request (see return of method)
Root<Question> root = cq.from(Question.class);
// Join happens based on respective attribute within root
root.join(Question_.USER_ANSWER);
// The following ifs are not the nicest solution.
// The ifs check what comparator String contains and adds respective where clause to query
// This .where() is like WHERE in SQL
if("==".equals(comparator)) {
cq.where(cb.equal(root.get(attribute), value));
}
if(">".equals(comparator)) {
cq.where(cb.gt(root.get(attribute), value));
}
if(">=".equals(comparator)) {
cq.where(cb.ge(root.get(attribute), value));
}
if("<".equals(comparator)) {
cq.where(cb.lt(root.get(attribute), value));
}
if("<=".equals(comparator)) {
cq.where(cb.le(root.get(attribute), value));
}
// Finally, query gets created and result collected and returned as List
// Hint for READ_ONLY is added as lists are often just for read and performance is better.
return entityManager.createQuery(cq).setHint(QueryHints.READ_ONLY, true).getResultList();
}
}
I have a Spring Boot App with Spring Data JPA with hibernate and MySQL as the data store.
I have 3 layers in my application:
API Service
Application Service
Domain Service ( with Repository)
The role of Application Service is to convert hibernate-backed POJOs to DTOs given some business logic.
POJO
SchoolClass.java
#Column
Long id;
#Column
String name;
#OneToMany(fetch = FetchType.LAZY, mappedBy = "schoolClass")
List<Book> books;
#OneToMany(fetch = FetchType.LAZY, mappedBy = "schoolClass")
List<Student> students;
#OneToMany(fetch = FetchType.LAZY, mappedBy = "schoolClass")
List<Schedule> schedules;
Domain Service - My transaction boundary is at the Domain Service layer.
SchoolClassService.java
#Autowired
private SchoolClassRepository repository;
#Transactional(readOnly = true)
public SchoolClass getClassById(Long id) {
return repository.findById(id);
}
Application Service
SchoolClassAppService.java
#Autowired
private SchoolClassService domainService;
public SchoolClassDto getClassById(Long id) {
SchoolClass schoolClass = domainService.getClassById(id);
// convert POJO to DTO;
return SchoolClassDto;
}
My problem is that at times the child entities on SchoolClass are empty when I try to access them in SchoolClassAppService. Not all of them, but out of the three, two would work fine but the third one would be empty. I tried to mark the children lists to be eagerly fetched, but apparently only two collections can be eagerly fetched before Hibernate starts throwing exceptions and it also does not sound like good practice to always load all the objects. I do not get LazyInitializationException, just the list is empty.
I have tried to just call the getter on all lists in the domain service method before returning it just to load all data for the POJO but that does not seem like a clean practice.
Are there any patterns available which keep the transaction boundaries as close to the persistence layer as possible while still make it viable to process the data even after the transaction has been closed?
Not sure why your collections are sometimes empty, but maybe that just how the data is?
I created Blaze-Persistence Entity Views for exactly that use case. You essentially define DTOs for JPA entities as interfaces and apply them on a query. It supports mapping nested DTOs, collection etc., essentially everything you'd expect and on top of that, it will improve your query performance as it will generate queries fetching just the data that you actually require for the DTOs.
The entity views for your example could look like this
#EntityView(SchoolClass.class)
interface SchoolClassDto {
String getName();
List<BookDto> getBooks();
}
#EntityView(Book.class)
interface BookDto {
// Whatever data you need from Book
}
Querying could look like this
List<SchoolClassDto> dtos = entityViewManager.applySetting(
EntityViewSetting.create(SchoolClassDto.class),
criteriaBuilderFactory.create(em, SchoolClass.class)
).getResultList();
Just keep in mind that DTOs shouldn't just be copies your entities but should be designed to fit your specific use case.
I have a many-to-many relation with an additional column in the link table. I've configured it in a way that the owning side fetches children eager (so I don't get LazyInitializationException) and in the opposite direction it is lazy. This works.
I now wanted to fine-tune the transactions (before there was just #Transactional on class level of DAO and Service classes. I set method getById to readOnly = true:
#Transactional(readOnly = true)
public Compound getById(Long id) {
return compoundDAO.getById(id);
}
After this change I get a LazyInitializationException in following snippet:
Compound compound = compoundService.getById(6L);
Structure structure = compound.getComposition().get(0).getStructure();
System.out.println("StructureId: "+ structure.getId()); // LazyInitializationException
If I remove (readOnly = true) this works! Can anyone explain this behavior? I use Spring + Hibernate. Kind of confusing as I don't see any reason why this should affect which data is loaded?
EDIT:
Snippets of relationship definitions. This is a many-to-many with a column in the link table.
Owning side (eg Compound contains Structures):
#OneToMany(fetch = FetchType.EAGER, mappedBy = "pk.compound",
cascade = CascadeType.ALL, orphanRemoval = true)
#OrderBy("pk.structure.id ASC")
private List<CompoundComposition> composition = new ArrayList<>();
Belongs to side:
#OneToMany(fetch = FetchType.LAZY, mappedBy = "pk.structure",
cascade = CascadeType.ALL)
#OrderBy("pk.compound.id ASC")
private List<CompoundComposition> occurence;
Many-To-One in #Embeddable ID class
#ManyToOne(fetch = FetchType.LAZY, cascade = CascadeType.ALL)
public Compound getCompound() {
return compound;
}
#ManyToOne(fetch = FetchType.LAZY, cascade = CascadeType.ALL)
public Structure getStructure() {
return structure;
}
EDIT 2:
Stack Trace
org.hibernate.LazyInitializationException: could not initialize proxy - no Session
at org.hibernate.proxy.AbstractLazyInitializer.initialize(AbstractLazyInitializer.java:165) ~[hibernate-core-4.1.7.Final.jar:4.1.7.Final]
at org.hibernate.proxy.AbstractLazyInitializer.getImplementation(AbstractLazyInitializer.java:272) ~[hibernate-core-4.1.7.Final.jar:4.1.7.Final]
at org.hibernate.proxy.pojo.javassist.JavassistLazyInitializer.invoke(JavassistLazyInitializer.java:185) ~[hibernate-core-4.1.7.Final.jar:4.1.7.Final]
at org.bitbucket.myName.myApp.entity.Structure_$$_javassist_0.getId(Structure_$$_javassist_0.java) ~[classes/:na]
at org.bitbucket.myName.myApp.App.main(App.java:31) ~[classes/:na]
EDIT 3:
Also see my comment:
Log is very different with readOnly and it is missing the part were the relations are loaded, eg. some selects are missing in the log.
EDIT 4:
So I tired with a basic DriverManagerDataSource and no Connection pool. The issue is exactly the same. For me looks like an issue in Hibernate.
This is just wow. I'm starting to understand why some people hate ORMs...Just feels like I'm constantly having to spend hours to solve a weird issue and the solution is a very specific set of annotations + some code to work around the limitations of said annotations.
First to why this happens (why meaning with which annotations, but not in terms of making logical sense, which is the actual problem here as using common-sense is useless. Only trial and error helps). In the owning side, in #OneToMany I have orphanRemoval = true (which I have found out is required for consistency. one would think database constraints should handle that...just one of the many things that can drive you crazy.). It seems that if the transaction is not read-only, then this setting leads to some data being fetched even so its lazy, namely here:
#ManyToOne(fetch = FetchType.LAZY, cascade = CascadeType.ALL)
public Structure getStructure() {
return structure;
}
In a read-only transaction, this fetching does not happen. I would guess because if you can't change anything you will also not have to remove orphans and hence any data that the logic behind this setting requires is not needed in a read-only tx.
So the obvious solution would be in above relation to change to FetchType.EAGER. Wrong! If you do that you will not be able to update the owning side (Compound) using session.merge. This will lead to a StackOverFlowError.
The real solution was actually already mentioned. Just leave the config as is but explicitly load the desired relations in the Service layer:
#Transactional(readOnly = true)
#Override
public Compound getById(Long id) {
Compound compound = compoundDAO.getById(id);
for (CompoundComposition composition : compound.getComposition()){
Hibernate.initialize(composition.getStructure());
}
return compound;
}
I admit I'm tending to fall in the premature optimization trap. This doesn't look very efficient and also seems to break how SQL works in the first place. But then I'm in the lucky position that in most cases CompoundComposition will contain only 1 or 2 elements.
Perhaps you could put
value.getComposition().get(i).getStructure();
in the body of the getById() method, so that the lazy loading happens within the transaction. I realize in this case you'd have to loop over i which might be inconvenient.
Two things :-
Lazy fetch works on Collections Interface. Since ...
#ManyToOne(fetch = FetchType.LAZY, cascade = CascadeType.ALL)
public Structure getStructure() {
return structure;
}
... this is not a collection interface (like List<Structure> would have been), it will be fetched in Eager fetch mode.
Make service method as transactional. It seems that after fetching from the dao layer, your structure is detached with NEVER flush mode. This is the underlying ORM issue I guess.