RE: tapestry-hibernate integration problem

2006-04-17 Thread Schulte Marcus
Why not use a separate session (same sessionfactory) for your logging? > -Original Message- > From: James Carman [mailto:[EMAIL PROTECTED] > Sent: Thursday, April 13, 2006 6:07 PM > To: 'Tapestry users' > Subject: RE: tapestry-hibernate integration problem > &

RE: tapestry-hibernate integration problem

2006-04-13 Thread James Carman
-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Thursday, April 13, 2006 8:55 AM To: 'Tapestry users' Subject: RE: tapestry-hibernate integration problem I'd just call it a feature ;) > -Original Message- > From: James Carman [mailto:[EMAIL

RE: tapestry-hibernate integration problem

2006-04-13 Thread James Carman
So, what do you suggest? -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Thursday, April 13, 2006 10:59 AM To: 'Tapestry users' Subject: RE: tapestry-hibernate integration problem You're right it's not a bug in the framework. I didn't

RE: tapestry-hibernate integration problem

2006-04-13 Thread Schulte Marcus
-Original Message- > From: Jean-Francois Poilpret [mailto:[EMAIL PROTECTED] > Sent: Thursday, April 13, 2006 4:25 PM > To: 'Tapestry users' > Subject: RE: tapestry-hibernate integration problem > > > Marcus, > > The problem you describe about a developer

RE: tapestry-hibernate integration problem

2006-04-13 Thread Jean-Francois Poilpret
1:15 PM To: 'Tapestry users' Subject: RE: tapestry-hibernate integration problem I currently think that declarative ta-demarcation at the service-facade in a typical web-app with hibernate is just plain wrong. Because it pretends to hide something (the hibernate session and its lifecycle) which

RE: tapestry-hibernate integration problem

2006-04-13 Thread Jean-Francois Poilpret
EMAIL PROTECTED] On Behalf Of Mark Reynolds Sent: Thursday, April 13, 2006 9:51 AM To: tapestry-user@jakarta.apache.org Subject: Re: tapestry-hibernate integration problem The problem I experienced using hivetranse and the "open session in view" was this: 1) execute one transactional meth

RE: tapestry-hibernate integration problem

2006-04-13 Thread Schulte Marcus
I'd just call it a feature ;) > -Original Message- > From: James Carman [mailto:[EMAIL PROTECTED] > Sent: Thursday, April 13, 2006 1:40 PM > To: 'Tapestry users' > Subject: RE: tapestry-hibernate integration problem > > > You're right. I

RE: tapestry-hibernate integration problem

2006-04-13 Thread James Carman
But, maybe there was an active transaction. Hm. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Thursday, April 13, 2006 2:20 AM To: 'Tapestry users' Subject: RE: tapestry-hibernate integration problem > -Original Message- > From: J

RE: tapestry-hibernate integration problem

2006-04-13 Thread James Carman
Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Thursday, April 13, 2006 2:15 AM To: 'Tapestry users' Subject: RE: tapestry-hibernate integration problem I currently think that declarative ta-demarcation at the service-facade in a typical web-app with hibernate is

RE: tapestry-hibernate integration problem

2006-04-12 Thread Schulte Marcus
> -Original Message- > From: James Carman [mailto:[EMAIL PROTECTED] > Sent: Thursday, April 13, 2006 12:42 AM > To: 'Tapestry users' > Subject: RE: tapestry-hibernate integration problem > > > Anything which really wants to update the persistent

RE: tapestry-hibernate integration problem

2006-04-12 Thread Schulte Marcus
rsday, April 13, 2006 4:51 AM > To: tapestry-user@jakarta.apache.org > Subject: Re: tapestry-hibernate integration problem > > > The problem I experienced using hivetranse and the "open session in > view" was this: > > 1) execute one transactional method,

Re: tapestry-hibernate integration problem

2006-04-12 Thread Nick Faiz
Mark, I havent noticed this yet. How did you first prove the existence of this problem? I will keep my eyes open for it. Nick On 13/04/2006, at 12:50 PM, Mark Reynolds wrote: The problem I experienced using hivetranse and the "open session in view" was this: 1) execute one transaction

Re: tapestry-hibernate integration problem

2006-04-12 Thread Mark Reynolds
The problem I experienced using hivetranse and the "open session in view" was this: 1) execute one transactional method, transaction is committed when the method returns and the hibernate session is kept open 2) execute another transactional method, hivetranse opens a second hibernate session,

Re: tapestry-hibernate integration problem

2006-04-12 Thread Nick Faiz
Apologies - I see now that there have been plenty of replies to the original post, including ones mentioning HiveUtils! Cheers, Nick On 13/04/2006, at 10:39 AM, Nick Faiz wrote: Just a note that you dont need to write code to control the Hibernate session & transaction lifecycles if you use

Re: tapestry-hibernate integration problem

2006-04-12 Thread Nick Faiz
Just a note that you dont need to write code to control the Hibernate session & transaction lifecycles if you use HiveUtils (aka Hivetranse). http://sourceforge.net/projects/hivetranse/ In my current project, I configured the hibernate session as a hivemind component, which is injected into

RE: tapestry-hibernate integration problem

2006-04-12 Thread James Carman
;s just "object soup." -Original Message- From: Pedro Viegas [mailto:[EMAIL PROTECTED] Sent: Wednesday, April 12, 2006 7:22 PM To: Tapestry users Subject: Re: tapestry-hibernate integration problem Check out this page with the Hivemind Utilities. It can help you set up the services yo

Re: tapestry-hibernate integration problem

2006-04-12 Thread Pedro Viegas
[EMAIL PROTECTED] On Behalf Of Mark Reynolds > Sent: Wednesday, April 12, 2006 6:01 PM > To: tapestry-user@jakarta.apache.org > Subject: Re: tapestry-hibernate integration problem > > So how does transaction demarcation work in this common tapestry sequence: > > 1) in page

RE: tapestry-hibernate integration problem

2006-04-12 Thread James Carman
-user@jakarta.apache.org Subject: Re: tapestry-hibernate integration problem So how does transaction demarcation work in this common tapestry sequence: 1) in pageBeginRender(), call a service method to load the persisted object and set it as a simple page property 2) tapestry form fields components

Re: tapestry-hibernate integration problem

2006-04-12 Thread Mark Reynolds
Of Sam Gendler Sent: Wednesday, April 12, 2006 11:26 AM To: Tapestry users Subject: Re: tapestry-hibernate integration problem I use the standard OpenSessionInView filter, which keep s a session open throughout the processing of a single request. As for transaction semantics, those are kept entirely in the

RE: tapestry-hibernate integration problem

2006-04-12 Thread James Carman
-- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Sam Gendler Sent: Wednesday, April 12, 2006 11:26 AM To: Tapestry users Subject: Re: tapestry-hibernate integration problem I use the standard OpenSessionInView filter, which keep s a session open throughout the processing of a single r

Re: tapestry-hibernate integration problem

2006-04-12 Thread Sam Gendler
I use the standard OpenSessionInView filter, which keep s a session open throughout the processing of a single request. As for transaction semantics, those are kept entirely in the service layer. The UI layer is completely oblivious to transaction semantiks precisely to avoid the problem you are

RE: tapestry-hibernate integration problem

2006-04-12 Thread James Carman
I wouldn't inject sessions into your pages directly. Dealing with the persistence store should be abstracted away from the view layer using a DAO/Repository layer. If you want an example of how to do that, you can download my example application (via SVN of course) from: http://www.carmanconsu

RE: tapestry-hibernate integration problem

2006-04-12 Thread Schulte Marcus
you should try an existing integration and adapt that if it doesn't fit your needs. the ones I know are: - Appfuse/Equinox - hivetranse (http://hivetranse.sf.net) - cognition (http://dev.thelabllc.com/cognition/) - honeycomb (http://honeycomb.javaforge.com) I use honeycomb (mostly because