[ http://jira.jboss.com/jira/browse/JBSEAM-1969?page=all ] Dan Allen updated JBSEAM-1969: ------------------------------ Attachment: JBSEAM-1969-v1.txt This problem is actually really simple to fix. Direct access to the entityClass private field just needs to be replaced with calls to the getter, as follows: getEntityClass() instead of entityClass However, there is one question. The @Create method calls getEntityClass(), which must be non-null and establishes the value of the entityClass private field. That begs the question, should subclasses be forced to call setEntityClass() in the getEntityClass() override? I am inclined to say that we should follow good OO principles and access the field using the getter method to support overrides such as this. However, the same problem will come up if you try to override getId(). I think in that case it is okay to enforce the call to setId() since that is a very special property. -- 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-1969?page=comments#action_12380090 ] Pete Muir commented on JBSEAM-1969: ----------------------------------- The patch is fine. I don't understand the problem. -- 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-1969?page=all ] Dan Allen updated JBSEAM-1969: ------------------------------ Attachment: HomeTest.java -- 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-1969?page=comments#action_12380145 ] Pete Muir commented on JBSEAM-1969: ----------------------------------- Sure, I understand the bug. I don't understand what you are talking about here "However, there is one question. The @Create method calls getEntityClass(), which must be non-null and establishes the value of the entityClass private field. That begs the question, should subclasses be forced to call setEntityClass() in the getEntityClass() override?" and "However, the same problem will come up if you try to override getId(). I think in that case it is okay to enforce the call to setId() since that is a very special property." -- 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-1969?page=comments#action_12380206 ] Dan Allen commented on JBSEAM-1969: ----------------------------------- The comment about this.id leading to NullPointerException inaccurately reflects what would happen. It will actually just lead to broken or lost functionality (missing dirty checks, etc) -- 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-1969?page=comments#action_12380222 ] Pete Muir commented on JBSEAM-1969: ----------------------------------- 1. Yes, this is what the bug is all about and the patch fixes 2. How is this.id used all over the place in Home? I see it used in 3 places: assignId, setId and getId. All of which are completely correct and allow overriding of getId. No, we certainly shouldn't make getId final. A large point of the framework is to allow you to use it as a base overriding where necessary *including* getId() (the ref docs even show doing this). And yes, if you override getId you are of course responsible for dirty checking on the field. -- 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-1969?page=all ] Dan Allen resolved JBSEAM-1969. ------------------------------- Fix Version/s: 2.0.0.CR2 (was: 2.0.1.GA) Resolution: Done commit comment: use getEntityClass() accessor rather than private field in base class to avoid NullPointerExceptions from being thrown if the developer overrides getEntityClass() but does not call setEntityClass() to establish the value of the private field. -- 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