GWT RequestFactory: updates not propagated to cascaded object - gwt

I have an entity Company with a referenced object ItemVersion and I use JPA (eclipselink) as persistence layer. A code extract is given here:
#Entity
public class Company{
private String instance;
#OneToOne(cascade = CascadeType.ALL)
#JoinColumn(name = "fk_item_version_id")
private ItemVersion itemVersion;
}
#Entity
public class ItemVersion{
private String comment;
...
}
I can create a Company object and persist it. I can also find the new object, update the attribute "instance" and persist it and everything works fine.
When I change the attribute "comment" of the referenced ItemVersion object, this change is not stored on the server side.
The create/update test code looks like:
final EventBus eventBus = new SimpleEventBus();
final AftdRequestFactory requestFactory = GWT.create(AftdRequestFactory.class);
requestFactory.initialize(eventBus);
final CompanyRequest request = requestFactory.companyRequest();
final CompanyProxy newCompany = request.create(CompanyProxy.class);
newCompany.setInstance("1");
ItemVersionProxy newVersion = request.create(ItemVersionProxy.class);
newVersion.setComment("first comment");
newCompany.setItemVersion(newVersion);
request.persist().using(newCompany).fire(new Receiver<Void>() {
#Override
public void onSuccess(Void arg0) {
final CompanyRequest request2 = requestFactory.companyRequest();
Request<CompanyProxy> p = request2.findCompany(1L).with("itemVersion");
p.to(new Receiver<CompanyProxy>() {
#Override
public void onSuccess(CompanyProxy response) {
final CompanyRequest request3 = requestFactory.companyRequest();
final CompanyProxy editableCompany2 = request3.edit(response);
editableCompany2.setInstance("2");
editableCompany2.getItemVersion().setVersionNumber(2);
request3.persist().using(editableCompany2).fire(new Receiver<Void>() {
#Override
public void onSuccess(Void arg0) {
// persist company version
System.out.println("company updated");
However, the update for "instance" and "comment" goes over the wire (checked wich wireshark between client and server), but in the persist method of Company, the referenced ItemVersion object and its "comment" attribute is not updated while "instance" is updated and therefore the old comment is stored.
The persist method of Company looks like:
public void persist() throws PersistenceException {
EntityManager em = emf.createEntityManager();
EntityTransaction tx = null;
try {
tx = em.getTransaction();
tx.begin();
Company existingEntity = findCompany(getId());
if (existingEntity == null) {
em.persist(this);
} else {
setId(existingEntity.getId());
em.merge(this);
}
tx.commit();
} catch (RuntimeException e) {
tx.rollback();
throw e;
}
The work around with a search and depending on the search result the em.persist or em.merge is necessary, because a simple persist does not store any updates.

You MUST use the open-session-in-view/session-per-request pattern (i.e. sharing the same EntityManager instance for the lifetime of the request) with RequestFactory; see https://code.google.com/p/google-web-toolkit/issues/detail?id=7827 for details.

Related

Transaction is required to perform this operation (either use a transaction or extended persistence context) while persisting Entity

I am trying to develop a class that runs at specific intervals and performs some DB modifications.
the code I have managed to run at a specific interval, retrieve records from the DB, but when I want to commit changes to the DB I get the following error.
WFLYEE0110: Failed to run scheduled task: javax.persistence.TransactionRequiredException: WFLYJPA0060: Transaction is required to perform this operation (either use a transaction or extended persistence context)
is #ApplicationScoped allowed to create transactions?
Thanks!
#ApplicationScoped
#ActivateRequestContext
public class TaskRunner {
#PersistenceContext(type = PersistenceContextType.EXTENDED)
EntityManager em;
#Resource private ManagedScheduledExecutorService scheduler;
private ScheduledFuture<?> TaskRunnerScheduler;
private boolean initialized = false;
private void init(#Observes #Initialized(ApplicationScoped.class) Object init) {
if (initialized) return;
initialized = true;
try {
// Execute at startup
TaskRunner = scheduler.schedule(this::runSchedule, getSchedule());
} catch (Throwable throwable) {
}
}
#Transactional
private void runSchedule() {
//retrieve db records
//make changes and commit
//sample
//em.persist(someEntity)
}
private Trigger getSchedule() {
return new Trigger() {
#Override
public Date getNextRunTime(LastExecution lastExecutionInfo, Date taskScheduledTime) {
return Date.from(
ZonedDateTime.now().withSecond(0).withNano(0).plusHours("4").toInstant());
}
#Override
public boolean skipRun(LastExecution lastExecutionInfo, Date scheduledRunTime)
{return false;}};
}
}
Transactions are started via Interceptors.
When you call a method of a bean from inside that bean, the method-call is not intercepted and no Transaction can get started.
You need another Bean and persist in there
#RequestScoped
#Transactional(value = TxType.REQUIRES_NEW)
public class SomeOtherBean{
#PersistenceContext(type = PersistenceContextType.EXTENDED)
EntityManager em;
public void doSomething(){
//retrieve db records
//make changes and commit
//sample
//em.persist(someEntity)
}
}
then you can Inject that bean in your TaskRunner
#ApplicationScoped
#ActivateRequestContext
public class TaskRunner {
#Inject
SomeOtherBean someBean;
...
private void runSchedule() {
someBean.doSomething()
}
}

one transaction for multiple contexts in integration tests on TestInitialize

I am writing integration tests and I want to use transaction scope.
We use EF and Repositories with Contexts.
If I have one Repository and once Context then it would look like this:
[TestInitialize]
public void RuleEngineTestsStart() {
customContext = new CustomContext();
transaction = customContext.Database.BeginTransaction();
repo = new CustomRepository(customContext);
// I need to make this context to work in the same transaction as above
anotherContext = new AnotherContext();
anotherRepo = new AnotherRepository(anotherContext);
}
At the end of tests (TestCleanup) I would like to transaction.Rollback(); everything.
I want to have the same transaction for all repositories that work with different contexts, is it possible? How to create transaction and 'send' it to all three contexts?
Please, to do not to use one Context for all repositories, it is not possible due to reasons (we want to have each context with its own DbSets later to be used within microservices).
Edit
In comments I was asked to include more code, however, I think is not necessary to answer my question.
customContext = new CustomContext();
repo = new CustomRepository(customContext);
customContext2 = new CustomContext2();
otherRepository = new CustomRepository2(customContext2);
// class to be tested needs both repositories
ToBeTestedClass cl = new ToBeTestedClass(customRepository, otherRepository);
// "BASE" interface
public interface IRepository<TEntity> where TEntity : class
{
TEntity GetById(long id);
IEnumerable<TEntity> GetByFilter(Expression<Func<TEntity, bool>> predicate);
TEntity GetSingleByFilter(Expression<Func<TEntity, bool>> filter);
void Insert(TEntity entity);
void Delete(long id);
void Update(TEntity entity);
...
}
// BASE CLASS
public class Repository<TEntity> : IRepository<TEntity> where TEntity : class
{
protected readonly DbContext _context;
protected readonly DbSet<TEntity> _dbSet;
public Repository(ColldeskDbContext context)
{
_context = context;
_dbSet = context.Set<TEntity>();
}
// GetSingle, GetAll, Insert, Update etc.
}
// CustomRepository (other Repositories are similar, with custom methods)
public interface ICustomRepository : IRepository<CusotmData>
{
// some specific methods that are not in Base class
}
public class CustomRepository: Repository<CustomData>, ICustomRepository
{
public CustomRepository(CustomContext context) : base(context)
{
}
// custom methods that are specific for given context
}
// Contexts - each context consists of its one DbSets
Don't use dbContext.SaveChanges() in your repositories. Use ONE dbContext when creating repositories. Sample:
using ( var db = new YourDbContext() )
{
// Create and begin transaction
using ( var transaction = db.Database.BeginTransaction() )
{
try
{
// ONE dbContext for all repositories
var firstRepo = new Custom1Repository(db);
var secondRepo = new Custom2Repository(db);
City city = new City { Description = "My city" };
Street street = new Street { Description = "My street", City = city};
firstRepo.Insert(city);
secondRepo.Insert(street);
// Save all your changes and after that commit transaction
db.SaveChanges();
transaction.Commit();
}
catch ( Exception ec)
{
transaction.Rollback();
}
}
}
Doing like this your repositories becomes just wrappers over DbSet<TEntity>
I have figured out that I can simply use TransactionScope like this:
private TransactionScope _scope;
[TestInitialize]
public void TestInitialize()
{
_scope = new TransactionScope();
}
[TestCleanup]
public void TestCleanup()
{
_scope.Dispose();
}
And then each Context would be running within this TransactionScope.

Why Lazy Collections do not work with JavaFX getters / setters?

I experienced poor performance when using em.find(entity, primaryKey).
The reason seems to be that em.find() will also load entity collections, that are annotated with FetchType.LAZY.
This small test case illustrates what I mean:
public class OriginEntityTest4 {
[..]
#Test
public void test() throws Exception {
final OriginEntity oe = new OriginEntity("o");
final ReferencePeakEntity rpe = new ReferencePeakEntity();
oe.getReferencePeaks().add(rpe);
DatabaseAccess.onEntityManager(em -> {
em.persist(oe);
em.persist(rpe);
});
System.out.println(rpe.getEntityId());
DatabaseAccess.onEntityManager(em -> {
em.find(OriginEntity.class, oe.getEntityId());
});
}
}
#Access(AccessType.PROPERTY)
#Entity(name = "Origin")
public class OriginEntity extends NamedEntity {
[..]
private final ListProperty<ReferencePeakEntity> referencePeaks =
referencePeaks =
new SimpleListProperty<>(FXCollections.observableArrayList(ReferencePeakEntity.extractor()));
#Override
#OneToMany(mappedBy = "origin", fetch = FetchType.LAZY)
public final List<ReferencePeakEntity> getReferencePeaks() {
return this.referencePeaksProperty().get();
}
public final void setReferencePeaks(final List<ReferencePeakEntity> referencePeaks) {
this.referencePeaksProperty().setAll(referencePeaks);
}
}
I cannot find any documentation on that, my question is basically how can I prevent the EntityManager from loading the lazy collection?
Why I need em.find()?
I use the following method to decide whether I need to persist a new entity or update an existing one.
public static void mergeOrPersistWithinTransaction(final EntityManager em, final XIdentEntity entity) {
final XIdentEntity dbEntity = em.find(XIdentEntity.class, entity.getEntityId());
if (dbEntity == null) {
em.persist(entity);
} else {
em.merge(entity);
}
}
Note that OriginEntity is a JavaFX bean, where getter and setter delegate to a ListProperty.
Because FetchType.LAZY is only a hint. Depending on the implementation and how you configured your entity it will be able to do it or not.
Not an answer to titles question but maybe to your problem.
You can use also em.getReference(entityClass, primaryKey) in this case. It should be more efficient in your case since it just gets a reference to possibly existing entity.
See When to use EntityManager.find() vs EntityManager.getReference()
On the other hand i think your check is perhaps not needed. You could just persist or merge without check?
See JPA EntityManager: Why use persist() over merge()?

How to implement Rest web service using spring 3?

I have a Library application which is already implemented in spring MVC.
I need to use ReST web services for the same application using spring 3.
I have a Controller class I want is to be as a RestFul webService
#Controller #SessionAttributes("category")
public class CategoryController {
private static final Log log = LogFactory.getLog(CategoryController.class);
#Autowired
private CategoryService categoryService;
#Autowired
private ItemService itemService;
#RequestMapping("/category/categoryList.htm")
public ModelAndView list(HttpServletRequest request,
HttpServletResponse response) throws Exception {
List<Category> list = categoryService.getAllMainCategories();
Map map = new HashMap();
map.put("categoryList", list);
map.put("category", new Category());
return new ModelAndView("categoryList", map);
}
#RequestMapping(method = RequestMethod.POST, value = "/category/save.htm")
public String save(HttpServletRequest request,
HttpServletResponse response, Category command) throws Exception {
log.debug("save method called" + command);
Category category = (Category) command;
System.out.println(category);
categoryService.saveCategory(category);
return "redirect:/category/categoryList.htm";
}
#RequestMapping("/category/edit.htm")
public String edit(#RequestParam String id, ModelMap model)
throws Exception {
log.debug("edit method called :" + id);
log.debug(Long.parseLong(id));
Category cat = categoryService.getCategory(Long.parseLong(id));
model.put("categoryList", categoryService.getAllMainCategories());
model.put("category", cat);
return "categoryList";
}
#RequestMapping("/category/delete.htm")
public String remove(#RequestParam String id, ModelMap model)
throws Exception {
log.debug("remove method called " + id);
categoryService.deleteCategory(Long.parseLong(id));
return "redirect:/category/categoryList.htm";
}
#InitBinder
protected void initBinder(WebDataBinder binder) {
binder.registerCustomEditor(Category.class,
new PropertyEditorSupport() {
#Override
public void setAsText(String text) {
setValue(categoryService.getCategory(Long.valueOf(text)));
}
});
}
}
it is CategoryController class which add delete or update a category
ItemService and CategoryService are data sources
Category is a domain object having properties like id,name,description etc..,
How do I write a REST web service for this?
There's a simple example showing how in Barebones Spring. Check it out.

how to update an entity in Entity Framework 4 .NET

my code is something like this:
public class Program
{
[STAThread]
static void main()
{
DataAccessClass dal = new DataAccessClass();
List<Person> list = dal.GetPersons();
Person p = list[0];
p.LastName = "Changed!";
dal.Update(p);
}
}
public class DataAccessClass
{
public static List<Person> GetPersons()
{
MyDBEntities context = new MyDBEntities();
return context.Persons.ToList();
}
public void Update(Person p)
{
// what sould be written here?
}
}
now please tell me what should i write in the Update() method?
everything i write , encounters various exceptions.
(please pay attention that the data loaded is tracked , connected or something like that)
The problem is that your Person entities are still attached to context created in GetPersons. If you want to work with attached entities you have to use same context instance in both select and update operations. You have two choices to solve your problem.
1) Correctly handled attached entities
public class Program
{
[STAThread]
static void main()
{
using (DataAccessClass dal = new DataAccessClass())
{
List<Person> list = dal.GetPersons();
Person p = list[0];
p.LastName = "Changed!";
dal.Save();
}
}
}
public class DataAccessClass : IDisposable
{
private MyDBEntities _context = new MyDBEntities();
public List<Person> GetPersons()
{
return _context.Persons.ToList();
}
public void Save()
{
// Context tracks changes on your entities. You don't have to do anything. Simply call
// SaveChanges and all changes in all loaded entities will be done in DB.
_context.SaveChanges();
}
public void Dispose()
{
if (_context != null)
{
_context.Dispose();
_context = null;
}
}
}
2) Don't use attached entities
public class Program
{
[STAThread]
static void main()
{
DataAccessClass dal = new DataAccessClass())
List<Person> list = DataAccessClass.GetPersons();
Person p = list[0];
p.LastName = "Changed!";
dal.Update(p);
}
}
public class DataAccessClass
{
public static List<Person> GetPersons()
{
// Closing context will detach entities
using (MyDBEntities context = new MyDBEntities())
{
return context.Persons.ToList();
}
}
public void Update(Person p)
{
using (MyDBEntities context = new MyDBEntities())
{
context.Persons.Attach(p);
// Detached entities don't track changes so after attaching you have to say
// what changes have been done
context.ObjectStateManager.ChangeObjectState(p, System.Data.EntityState.Modified);
context.SaveChanges();
}
}
}
Taken from Employee Info Starter Kit, you can consider the code snippet as below:
public void UpdateEmployee(Employee updatedEmployee)
{
//attaching and making ready for parsistance
if (updatedEmployee.EntityState == EntityState.Detached)
_DatabaseContext.Employees.Attach(updatedEmployee);
_DatabaseContext.ObjectStateManager.ChangeObjectState(updatedEmployee, System.Data.EntityState.Modified);
_DatabaseContext.SaveChanges();
}
does not work when you have a property on entity which is a ConcurrencyToken.
At least for me. Because you then get a OptimisticConcurrencyException.
What i do (and i think this is not an optimum solution),
facts:
- I use a new context because of n-tier. So, the previous/original entity with its values are not known. Either you supplies the context with original and old (bah) or like me load original first prior to update:
T originalItem = sessionManager.Set().Single(x => x.ID == changedEntity.ID);
if(changedEntity.lastChangedDate != originalItem.lastChangedDate)
throw new OptimisticConcurrencyException(String.Format("Trying to update entity with lastChangedDate {0} using lastChangedDate {1}!", originalItem.lastChangedDate, changedEntity.lastChangedDate));
ObjectStateEntry state = sessionManager.ObjectStateManager.GetObjectStateEntry(originalItem);
state.ApplyCurrentValues(changedEntity);
state.ChangeState(System.Data.EntityState.Modified);
sessionManager.SaveChanges();
If you know something better, please let me know.
Atam