[ http://jira.jboss.com/jira/browse/JBSEAM-2751?page=comments#action_12403736 ] Norman Richards commented on JBSEAM-2751: ----------------------------------------- The workaround is to make sure you @In a Datamodel that you @Out from a conversational component. This will cause Seam to not manage the DataModel specially. I'm not sure what the correct thing to do here. To do any more would completely break the current behavior with respect to DataModel fields, however I'm really not sure we are doing the right thing. I can't see any case where the DataModel wouldn't be exposed outwardly, either with @Out or with a getter. And when it is, pulling the DataModel contents out like we do will cause the datamodel to be unusable. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JBSEAM-2751?page=all ] Norman Richards updated JBSEAM-2751: ------------------------------------ Comment: was deleted -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JBSEAM-2751?page=comments#action_12403928 ] Przemyslaw Jaskierski commented on JBSEAM-2751: ----------------------------------------------- Norman, Sorry for this clumsy comment, I made it linking those two issues. My original problem has been described in JBSEAM-1840: my List<HibernateEntity>, NOT annotated with @DataModel, is magically turned into List of nulls. Only workaround to prevent ManagedEntityIdentityInterceptor from nullyfing this List is adding "transient" to List's declaration (from the ManagedEntityIdentityInterceptor it appears that such a field is not processed). I've linked these two issues, because this problem hasn't been solved yet and it kinda return from time to time. And is very nasty to spot. Currently, because of it's nondeterminism, this is only bug that prevents me from feeling comfortably using Seam :) No automanaging magic should be counterintuitive to well established expectation that I'm controlling the state of my object :) This problem is comming over and over again, another one: JBSEAM-2684 -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JBSEAM-2751?page=comments#action_12403934 ] Przemyslaw Jaskierski commented on JBSEAM-2751: ----------------------------------------------- I cannot express myself in english as good as I want to :( Problem is described in JBSEAM-1840 Long running conversation backed by a JPDL pageflow. Backed by some conversational-scoped pojos. Pojos have fields like: class MyPojo { private List<@Entity> list; private @Entity entity; } 1. First step of flow initializes this list using seam-managed LRC. 2. process by the page flow N steps forward 3. go back using pageflow n<<N steps (still in conversation) 4. Instead of originally initialized list = [@Entity1, @Entity2...] with .size() == 10, I get list=[null, null...] with .size() == 10. But when declaring by pojo like this: class MyPojo { private TRANSIENT List<@Entity> list; private TRANSIENT @Entity entity; } this problem cease to appear. Why? Because (JBSEAM-1840) ManagedEntityIdentityInterceptor leaves TRANSIENT fields alone (ignore(Field) method). This is almost the case with any of the linked issues, so there is a real nasty problem with ManagedEntityIdentityInterceptor. There are test cases shown there, even a new one added by Dany to JBSEAM-1840: http://jira.jboss.com/jira/browse/JBSEAM-1840#action_12387051 I cannot describe this problem any better, Norman. If this has no value for you, feel free to ignore it. Regards, Przemek. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JBSEAM-2751?page=comments#action_12404028 ] Przemyslaw Jaskierski commented on JBSEAM-2751: ----------------------------------------------- Sure, thank you. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JBSEAM-2751?page=comments#action_12404242 ] Przemyslaw Jaskierski commented on JBSEAM-2751: ----------------------------------------------- Alex, please note that this ugly workaround is a timebomb waiting to hurt you badly in case of session passivation/clustering I guess... -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira