SessionFactory has a getCurrentSession function now, and you can define a policy if you dont like the existing. All you need is a servlet filter to commit & close the current session. It's very few lines of code saved by using Spring ORM.
Henrik "James Carman" <[EMAIL PROTECTED]> skrev i en meddelelse news:[EMAIL PROTECTED] > The main reason that I use the Spring stuff is for the transaction > demarcation stuff. Hibernate doesn't include anything "out-of-the-box" > that > allows you to do "nested" transactions (i.e. REQUIRES_NEW). You'd have to > roll that yourself and Spring has already done that for me. Although, I'm > looking into just implementing it on my own, since it's really not that > difficult to do (keep a stack around for your "current session"). > > -----Original Message----- > From: news [mailto:[EMAIL PROTECTED] On Behalf Of hv @ Fashion Content > Sent: Wednesday, August 16, 2006 3:35 PM > To: users@tapestry.apache.org > Subject: Re: tapernate question > > I had a similar concern with Trails, and made my own persistence service > that uses Hibernate directly. > With Hibernate 3.1 you don't really need the Spring ORM stuff. > > "Xiaoshu Wang" <[EMAIL PROTECTED]> skrev i en meddelelse > news:[EMAIL PROTECTED] >> Thanks, James. >> >> I am able to get the exception now. However, the exceptions seem to be >> wrapped into spring's DataAccessException. The original cause of >> HibernateException is gone. I tried to traverse the Exception track by >> getCause(), but only get the DataIntegrityViolationException and >> java.sql.BatchUpdateException. >> >> I am writing an application that has two fields that can violate the same >> contraints. The lost of HibernateException won't allow me to figure out >> the >> which field is duplicated. >> >> Xiaoshu >> >>> -----Original Message----- >>> From: James Carman [mailto:[EMAIL PROTECTED] >>> Sent: Wednesday, August 16, 2006 6:05 AM >>> To: 'Tapestry users' >>> Subject: RE: tapernate question >>> >>> If you want more fine-grained control over what's going on, >>> turn off transaction-per-request and use service methods with >>> transaction interceptors on them to achieve what you want. >>> The reason that you didn't get the exception is that the >>> transaction isn't committing until the end of the request and >>> that's when you'll get the unique constraint violation >>> exception. The exception presenter might not even catch >>> this, come to think of it. I actually changed the way >>> Tapernate is implemented in my local copy. I might need to >>> push that out to the rest of you. >>> >>> -----Original Message----- >>> From: Xiaoshu Wang [mailto:[EMAIL PROTECTED] >>> Sent: Tuesday, August 15, 2006 9:45 PM >>> To: 'Tapestry users' >>> Subject: RE: tapernate question >>> >>> I tried to put something like >>> >>> try { >>> getSession.update(message); >>> } catch (Throwable e) { >>> System.out.println("Catch it."); >>> } >>> >>> under the MessageDaoImpl.update(Message), but it seems not >>> doing anything. >>> >>> How to hook with ExceptionPresenter? >>> >>> Xiaoshu >>> >>> >>> > -----Original Message----- >>> > From: James Carman [mailto:[EMAIL PROTECTED] >>> > Sent: Tuesday, August 15, 2006 9:09 PM >>> > To: 'Tapestry users' >>> > Subject: RE: tapernate question >>> > >>> > You can put in a hook into the ExceptionPresenter or wrap your call >>> > with a try/catch block. The HibernateService class can be found at: >>> > >>> > http://svn.javaforge.com/svn/hivemind/hivemind-utils/trunk/src >>> /main/java/com >>> > /javaforge/hivemind/util/HiveMindService.java >>> > >>> > The username/password is anonymous/anon. >>> > >>> > >>> > -----Original Message----- >>> > From: Xiaoshu Wang [mailto:[EMAIL PROTECTED] >>> > Sent: Tuesday, August 15, 2006 9:03 PM >>> > To: users@tapestry.apache.org >>> > Subject: tapernate question >>> > >>> > Hi, I am playing around Tapernate, which I have a question >>> to ask. It >>> > seems that the exception thrown by a database query would >>> not be able >>> > to be catched at the application level. >>> > For isnstance, I justed added a unique constraints on the >>> > Message.value property. If I ever try to created a message >>> that has a >>> > duplicated value with an existing Message, it break the >>> application. >>> > I didn't find a way to catch the exception. Is this >>> expected behavior >>> > or not? >>> > >>> > Also, where I can find the source code for the HibernateService? >>> > >>> > Thanks, >>> > >>> > XW >>> > >>> > >>> > >>> --------------------------------------------------------------------- >>> > To unsubscribe, e-mail: [EMAIL PROTECTED] >>> > For additional commands, e-mail: [EMAIL PROTECTED] >>> > >>> > >>> > >>> > >>> --------------------------------------------------------------------- >>> > To unsubscribe, e-mail: [EMAIL PROTECTED] >>> > For additional commands, e-mail: [EMAIL PROTECTED] >>> > >>> > >>> > >>> >>> >>> --------------------------------------------------------------------- >>> To unsubscribe, e-mail: [EMAIL PROTECTED] >>> For additional commands, e-mail: [EMAIL PROTECTED] >>> >>> >>> >>> --------------------------------------------------------------------- >>> To unsubscribe, e-mail: [EMAIL PROTECTED] >>> For additional commands, e-mail: [EMAIL PROTECTED] >>> >>> >>> >> >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [EMAIL PROTECTED] >> For additional commands, e-mail: [EMAIL PROTECTED] >> >> > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
