[JBoss-user] [JBoss Seam] - Re: SeamExtendedManagedPersistencePhaseListener

2006-05-25 Thread g.tomassoni
Well, SEMPPL seems work fine to me in JBoss 4.0.4GA + Seam 20060520 + jsf-facelets 1.1.4. It has still a drawback, anyway: when an exception occurs in a page, it leaks the serving transaction which is left active in the thread and may eventually be propagated to other servlets. For the very

[JBoss-user] [JBoss Seam] - Re: SeamExtendedManagedPersistencePhaseListener

2006-05-25 Thread g.tomassoni
pbrewer_uk wrote : I do sometimes experience an exception that brings down the whole servlet I think this is not the same problem I'm experiencing: I get errors caused by transactions already active after the first exception. Please note that the first exception is due to a software mistake

[JBoss-user] [JBoss Seam] - Re: resume parent conversation in @End method

2006-05-16 Thread g.tomassoni
This is an interesting new feature. Do you think there would be any usefulness in enhancing it (or perhaps extending it) to the case in which one is popping and redirecting from the topmost conversation? I guess it could be usefull in many cases. In example, a logon issued by a login

[JBoss-user] [JBoss Seam] - Re: resume parent conversation in @End method

2006-05-12 Thread g.tomassoni
Great! Gonna try it! Thanks, View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3943242#3943242 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3943242 --- Using Tomcat but

[JBoss-user] [JBoss Seam] - Re: SeamExtendedManagedPersistencePhaseListener

2006-04-28 Thread g.tomassoni
I found a (very) temporary workaround to the transaction-still-there: don't use the SeamExtendedManagedPersistencePhaseListener. I had somewhere a servlet filter adding transactional behaviour to http requests. It basicly wraps a transaction begin/commit around an http request, rolling the

[JBoss-user] [JBoss Seam] - Re: SeamExtendedManagedPersistencePhaseListener

2006-04-28 Thread g.tomassoni
anonymous wrote : If you try to mix your own tx management with Seam tx management, you will break stuff. No wonder you have been having problems. Somehow, I'd the feeling you would reply this way... :) Gavin, I added the transactional filter to the request chain to use the SeamServletFilter,

[JBoss-user] [JBoss Seam] - Re: SeamExtendedManagedPersistencePhaseListener

2006-04-28 Thread g.tomassoni
anonymous wrote : As I read his thread, he did stop using SEMPPL and switched to the regular SeamPhaseListener + custom TX management because he could not resolve the exception. He was never simultaneously using SEMPPL + custom TX management. Right. View the original post :

[JBoss-user] [JBoss Seam] - Re: SeamExtendedManagedPersistencePhaseListener

2006-04-26 Thread g.tomassoni
What about the following? After restarting jboss and the app and accessing this page from the main one, the page is correctly shown but the request ends with a RollbackException, which mean someone marked rollback-only the transaction. Please note that no place in the involved application code

[JBoss-user] [JBoss Seam] - SeamExtendedManagedPersistencePhaseListener

2006-04-24 Thread g.tomassoni
I'm having a bad time with it. I'm developing a webapp in which I have conversations (and plan to use jBPM) and some custom servlet for handling images. The problem is that, sometime, my image servlet throws an exception by which I'm told that a UserTransaction is already active. The image

[JBoss-user] [JBoss Seam] - Re: SeamExtendedManagedPersistencePhaseListener

2006-04-24 Thread g.tomassoni
You're right: cold-deploying the app I don't get anymore an IllegalWhatever exception due to an already-started transaction from the non-seam servlet. But now I get the following when accessing a parametrized page (eg.: thepage.jsf?c=3) from the very same page but different parameter (eg: from

[JBoss-user] [JBoss Seam] - Re: SeamExtendedManagedPersistencePhaseListener

2006-04-24 Thread g.tomassoni
Oh, by the way: I'm now using JBoss-4.0.4-CR2 and jboss-seam-CVS20060423. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3939072#3939072 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3939072

[JBoss-user] [JBoss Seam] - Re: SeamExtendedManagedPersistencePhaseListener

2006-04-24 Thread g.tomassoni
Well, it may be. But why it happens on finalizing a component and not when injecting EMs? Also, if it is a classpath problem, I would expect to see it everytime, not in few, exact conditions. Isn't that there is a classloader problem somewhere? Like, say, the tomcat's thread pool is grown

[JBoss-user] [JBoss Seam] - Re: @In, @Out and @Factory. It's me or...

2006-04-15 Thread g.tomassoni
Almost: splitting the MyClass in two classes, one outjecting idCat and one outjecting currentCategory, as well as emulating the behaviour of @Factory through @Out(s) on getters solved the problem. It is not that elegant, but works. After changes I thought that retrieving idCat in

[JBoss-user] [JBoss Seam] - @In, @Out and @Factory. It's me or...

2006-04-08 Thread g.tomassoni
Dears, I have something like this in a Seam component of mines: | @Name(myComp) | @Scope(ScopeType.EVENT) | @Intercept(InterceptionType.ALWAYS) | public class MyComp { | @In(create = true) | private HibernateEntityManager theDatabase; | | | @RequestParameter |

[JBoss-user] [JBoss Seam] - Re: @In, @Out and @Factory. It's me or...

2006-04-08 Thread g.tomassoni
It is not that easy to post: it is based on a two-level template and two custom tags in a ui:repeat/c:forEach... What are you looking for? I'll post the needed fragment. This is the base page, however: | ?xml version=1.0 encoding=UTF-8? | !DOCTYPE composition PUBLIC -//W3C//DTD XHTML 1.0

[JBoss-user] [JBoss Seam] - Re: @In, @Out and @Factory. It's me or...

2006-04-08 Thread g.tomassoni
Well, context demarcation happens in component code, not in jsf page... However, actually I don't use it at all. I will for some pages, but not yet. @Out with @Factory always works to me when the outjected value is asked by jsf pages. It never works to me when it is used from inside the same

[JBoss-user] [JBoss Seam] - Re: @In, @Out and @Factory. It's me or...

2006-04-08 Thread g.tomassoni
Ok, my brain was elsewhere while writing my previous post... Now I got your point. I think there shouldn't be any difference in grabbing an attr at context or from component level: after all, in any case there is an underlying faces context triggering the (in my case cascaded) bijections...

[JBoss-user] [JBoss Seam] - Re: @In, @Out and @Factory. It's me or...

2006-04-08 Thread g.tomassoni
Of course, it is: | @RequestParameter | private String c; | Something went wrong in my previous cut'n'paste. The code I'm using is correct, however... View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3936015#3936015 Reply to the post :

[JBoss-user] [JBoss Seam] - Re: DataModel / Duplicated ID

2006-03-27 Thread g.tomassoni
[EMAIL PROTECTED] wrote : Yes, it does seem to fix the bug in all cases. You have to return an outcome and trigger a navigation rule from faces-config.xml. I have this problem, too. But I don't (yet) have any interceptor triggered: I'm actually prototipizing a site and have only navigation

[JBoss-user] [JBoss Seam] - Re: DataModel / Duplicated ID

2006-03-27 Thread g.tomassoni
[EMAIL PROTECTED] wrote : Guys this is NOTHING TO DO WITH SEAM Well, I didn't mind to call for an impeachment of Seam in this. Just, since the people using Seam is also using myfaces+facelets, chances are they are up to date with news about this issue... Did I do something wrong? View the

[JBoss-user] [JBoss Seam] - Re: DataModel / Duplicated ID

2006-03-27 Thread g.tomassoni
However, you were right, Gavin: upgrading the MyFaces core library (api+impl) to the 1.1.3-SNAPSHOT dated march 27, 2006, solved this issue at least for me. To whom it may concern, the MyFaces core's 1.1.3 nightly snapshots may be downloaded from http://cvs.apache.org/builds/myfaces/nightly/.