JBossDeveloper

For mail there is already a Seam 3 module for it which is scheduled for a full release when the full stack of Seam 3.1 is released.  In the meantime you can pick up a snapshot version for initial development locally.  If you have any problems with mail please contact the module lead, Cody Lerum.  See http://seamframework.org/Seam3/MailModule for details.

As for EntityHome, I don't believe there are any existing plans to implement equivalent functionality within Seam 3, as it is possible do the following now:

public abstract class PersistenceUtil implements Serializable
{
   private static final long serialVersionUID = -276417828563635020L;

   protected abstract EntityManager getEntityManager();

   protected <T> void create(final T entity)
   {
      getEntityManager().persist(entity);
   }

   protected <T> void delete(final T entity) throws NoResultException
   {
      getEntityManager().remove(entity);
   }

   protected <T> T deleteById(final Class<T> type, final Long id) throws NoResultException
   {
      T object = findById(type, id);
      delete(object);
      return object;
   }

   @SuppressWarnings("unchecked")
   protected <T> T findById(final Class<T> type, final Long id) throws NoResultException
   {
      Class<?> clazz = getObjectClass(type);
      T result = (T) getEntityManager().find(clazz, id);
      if (result == null)
      {
         throw new NoResultException("No object of type: " + type + " with ID: " + id);
      }
      return result;
   }

   protected <T> void save(final T entity)
   {
      getEntityManager().merge(entity);
   }

The above is a snippet of the PersistenceUtil class that is generated by Seam Forge when you specify that your project uses JPA.

If that isn't a good replacement for EntityHome, then please raise a jira at https://issues.jboss.org/browse/SEAMPERSIST.

Hi Ken,

By a full set of documentation I mean the availibility of guidance to start with Seam 3 just as with Seam 2 and additionally ; I REALLY started to appreciate Seam AFTER buying Dan's book (the other available books are just a waste of money!)
It would REALLY be helpful if there was some good migration information. I don't want a tool I want guidance.

Seam 2 suffered the same start-up problems, but I think there should be some lessons learned there. Personally I am not a real fan of seam-gen, because I started using maven from the first day. However you can't do full fledged Maven builds with Seam, I sincerely see it's more the other way around. You need a lot of tricks to do the full dev cycle.

I come from the time that console based programming was normal. I was really happy seeing IDE developments and I see the return of console programing (after 25 years we're back to the basics; faster but equally console based).

I'm back doing the number guessing examples and do some brain cracking reverse engineering.

Please don't have the idea that I am against things, but you're doing all the fancy stuff for a very small minority of wizz-kids. I belong to the overwhelming majority of people who like to have Seam-3 for Dummies. I just created a Blog which will contain my Seam 3 quest until fall this year. At the moment it has some useful content I will provide a link. It will be my contribution to something great, which not only needs technical support, but also better comercial support from Redhat.

Leo

Hi all,

the other thread I referred to:

Seam 3 remarks

The magic word there is productivity, and they're right. At the moment I can do an average complex Use-Case within a day with Seam 2. It will take months to be there with Seam 3. Is it worth the effort? What does my customers get? (fasetr time-to-market, cheaper ...)

Please, please let us have the Seam-2 JSF-2 duo for the time being..

Leo

Well, JEE6 / Seam 3 is a much more faster and produces less more errors then Seam 2...

I Think the major problem by porting seam 2 to JSF 2 is that there ar esome major changes in JSF 2.0, like the Scope behaviour. That is making it a big effort (may not expected before month) to integrate it. And I could really understand if they say, it is to much effort for this dieing technology.

And I don't understand why everybody wanna tell that it is so hard to get into Seam 3 / CDI. If you ahve worked with Seam 2 it couldn't be easier, because u know the complete concept behind it. I started workin with Seam 3 and loved it. Don't wanna go back to Seam 2, in major case of the performance and really easier injection handling in cdi.

Florian

Hello Ken,
Thanks for your reply.  We're using the entire Seam2 stack.  Obvious things that are missing for us is Drools support, JBPM (which doesn't appear even in the sandbox), Mail (although you address that above), and PDF and Excel generation.  I'm confident Seam 3 is ready for some people, it's just not ready for our applications.  The JSF2 issue is important for number of reasons:  1. Seam 2 was claimed to support it which affected our development direction; and 2. Support for JSF 1 by vendors like Primefaces is non-existent.  So, I think I'm stuck in the Seam2 world because Seam3 doesn't have everything Seam2 does, and support for Seam2 is waning.   I agree with Leo on the need and urgency of getting Seam2-JSF-2 in place.  Thanks again for your comments.

Robert I agree fully and was just about to post a similar thing. We have a number of Seam 2 apps and have also made use of almost every feature in the stack. Until similar features are released for Seam 3 we cannot even consider migrating any of our apps or even new applications as we need more that is available in Seam 3 and dont want to spend time developing custom integration when official modules are arriving slowly.

Also as you mention other vendors are quickly dropping support for JSF 1.2. Some of the issues we are starting to see are things like Richfaces 3 not working properly in IE9 and they have no intention of fixing it as Richfaces 4 is out but sadly this wont work with JSF 1.2!

Finally in the real commercial world companies simply cannot just stop developing their products and attempt to migrate to Seam 3. Real world existing applications need to continue to run in Seam 2 for a while and this is becoming increasingly difficult. Management in most companies wont even consider the risks of a major migration until there is a clear and defined migration path.

Hopefully this post isn't too blunt but I'm currently a bit frustrated at the sudden drop of Seam 2 and the lack of understanding why the community still needs this at least for a while. I would like nothing more than to start using Seam 3 tomorrow as it looks very interesting but sadly that is not a reality for me.

Craig.

Richfaces 3 will not be supported anymore I think, richfaces dev team is changed for 100%

and new is involved in richfaces 4 functionality.

Recently I attended JAX conf and jboss guys told me that they won't support Seam 3 + JSF 2.

All dev is concentrated under Jboass AS 7 and Seam3 + Weld.

So that's how things are going...

Sorry, won't support Seam 2 + JSF 2

Alexandr Areshchanka wrote on Jun 30, 2011 14:54:


Sorry, won't support Seam 2 + JSF 2

That is not true, I should know something about it ;-)

Regardless of what may or may not be happening with Seam 3 in the future in terms of additional modules etc. Us real world developers feel left out in the cold by the sudden drop of everything JSF 1.2 related and for all the fun and high levels of productivity we have experienced with Seam 2 we are left wondering if we have backed the wrong technology.

Well, I've downloaded and built Marek's experimental JSF2 source.  Just had to add a couple of repositories to get it to build, but so far so good. When I get to the point of migrating my stuff over to it, I'll start another thread.  As always, thank you Marek for doing this.  Can't tell you how much we appreciate this.

Marek Novotny wrote on Jun 30, 2011 17:04:

Alexandr Areshchanka wrote on Jun 30, 2011 14:54:


Sorry, won't support Seam 2 + JSF 2

That is not true, I should know something about it ;-)

May be I'm mistaken but 6 months past and no progress at all,
does jboss planning to make any efforts in that directions? No new people hired to develop Seam framework and no any announcements in community forum to involve people for committing new code...

I think many people would agree to spent 3-5 hours per week to migrate Seam 2 to JSF2 but it's only silence and empty promising at all.

I'm looking at Spring progress and see that it would more easear and reliable to use spring 3 + SWF plus it's all ready to use in production modules rather then wait for fresh jboss as 7\weld\seam3 stack...

Robert,
Could you please list the repositories you had to add to get JSF2 source to build? Thanks!

Russell Adcock wrote on Jul 01, 2011 07:58:


Robert,
Could you please list the repositories you had to add to get JSF2 source to build? Thanks!

This is good start (svn settings.xml ) for set up your local maven settings with the repository and as addition you have to set up java.net maven repository (http://download.java.net/maven/2/) to get jsf2 dependencies.

I'm currently using Maven2 for my Seam builds because PrimeFaces 1.1 requires a skinny war deployment, and I'm having issues with Maven 3 producing the required EAR.  In any case, I used Maven 3 for Marek's source, but it had some issues finding the Richfaces 3.3.3.Final jars and plugins.  I Googled them, and found repositories that had them.   I added the following to the head POM.  There are probably other locations, but these worked. 

The link in Marek's post above, requires a password.

<repositories>
     <repository>
          <id>maven.atlassian.com</id>
          <name>Atlassian Repository</name>
          <url>https://maven.atlassian.com/content/groups/public/</url>
     </repository>
</repositories>
     
<pluginRepositories>
     <pluginRepository>
          <id>maven.nuxeo.org</id>
          <name>Nuxeo Repository</name>
          <url>https://maven.nuxeo.org/nexus/content/groups/public/</url>
     </pluginRepository>
</pluginRepositories>