Does Hibernate throw and error if the result set is above a

SuperMan2000 · **Joined:** Fri Dec 15, 2006 5:49 pm **Posts:** 16

Does Hibernate throw and error if the result set is above a specific #
of records?

batmat · **Posted:** Thu Dec 21, 2006 1:05 pm

By default, no. It will try and load everything retrieved even until an OutOfMemoryError if there's a lot of results.

But maybe have a look at the session.setMaxResults(int) method.

SuperMan2000 · **Joined:** Fri Dec 15, 2006 5:49 pm **Posts:** 16

before I was running a single update as below and it worked fine:

Code:

    session.beginTransaction();
    /*
     int taskId = Integer.parseInt(request.getParameter("taskId"));
     String action_taken = request.getParameter("action_taken");

       WorkListErrors worklistErrors 
        = (WorkListErrors) session.load(WorkListErrors.class, new Integer(taskId));
        
        worklistErrors.setAction_taken(action_taken); 
        session.update(worklistErrors); 
        session.save(worklistErrors); 
        
        session.getTransaction().commit();   

but when I try an an update on multiple records it does work when I have a check box checked :

Code:

    session.beginTransaction();
    /*
     int taskId = Integer.parseInt(request.getParameter("taskId"));
     String action_taken = request.getParameter("action_taken");

       WorkListErrors worklistErrors 
        = (WorkListErrors) session.load(WorkListErrors.class, new Integer(taskId));
        
        worklistErrors.setAction_taken(action_taken); 
        session.update(worklistErrors); 
        session.save(worklistErrors); 
        
        session.getTransaction().commit();   


    session.beginTransaction();
     String[] tickedTaskId = request.getParameterValues("tickedTaskId");
    String[] taskId = request.getParameterValues("taskId");
     String[] action_taken = request.getParameterValues("action_taken");
     
     for(int i=0; i<tickedTaskId.length; i++) {

       for(int j = 0; j < taskId.length; j++) {
           
           if(tickedTaskId[i].equals(taskId[j])) {

               WorkListErrors worklistErrors 
               = (WorkListErrors) session.load(WorkListErrors.class, tickedTaskId[i]);
               
               worklistErrors.setAction_taken(action_taken[j]); 
               session.update(worklistErrors); 
               session.save(worklistErrors); 
               
               session.getTransaction().commit();   
           }
           
        }
     }     
    
    
     /*Close session */
      session.close();

batmat · **Posted:** Thu Dec 21, 2006 6:42 pm

SuperMan2000 wrote:

but when I try an an update on multiple records it does work when I have a check box checked :

I absolutely don't see what you're speaking about : "check box checked" :-/.

This code is quite bizarre:

Code:

session.update(worklistErrors);
session.save(worklistErrors);

updating, THEN saving? Are you sure you shouldn't just do one of these operations?

Moreover, you're doing a batch-processing. You should regularly flush() and clear() the cache as explained. I guess you should have a look in the reference documentation : http://www.hibernate.org/hib_docs/v3/re ... tch-update

SuperMan2000 · **Joined:** Fri Dec 15, 2006 5:49 pm **Posts:** 16

I though you require a save after and update...

so is the below correct

Code:

session.update(taskItem); 
session.getTransaction().commit();    
session.flush();
session.clear();
session.close(); 

andresgr · **Posted:** Fri Dec 22, 2006 1:36 pm

You are doing some things incorrectly (several commits, for example).

Just read this:

http://www.hibernate.org/hib_docs/v3/re ... tch-update

and do the same.

batmat · **Posted:** Fri Dec 22, 2006 5:49 pm

SuperMan2000 wrote:

I though you require a save after and update...

No, you don't.

Quote:

so is the below correct

Code:

session.update(taskItem); 
session.getTransaction().commit();    
session.flush();
session.clear();
session.close(); 

Yes, completely.

Even more to say: depending on your mapping, you should rarely need to ever call update(). In fact, calling save just says to Hibernate that the given instance must be saved in the db in the next flush(). At flush() time, Hibernate gets your id and see if it's null, for example (see the unsaved-value attribute semantics to understand this) and call a sequence to initialize it as a new INSERT. If it detects the id of your bean is not null, it then considers that it's an UPDATE that has to be called.