Well... reading the issue process slides once more, I
guess I was wrong. I am not sure I am happy about having just about anything
recorded in JIRA w/o letting the group at least notice its existence (and I do
not refer to a mail message).
Henning
My understanding was that issues are entered into the
JIRA (and given status "new") *before* they are discussed. When the TC
accepts them, then they can be given status "open", and if the TC does not
accept them, then they can be put in status "rejected".
When this is not correct, I can always delete the issue,
with apologies all round.
Ron
That issue was not even accepted,
right?
Henning
http://www.osoa.org/jira/browse/JAVA-5
TARGET: EJB Session Binding
Specification section 2.3.1 (section number from 1.0 version of the
spec)
DESCRIPTION:
In the discussion of the Ejb 2.x
client view in section 2.3.1 it says:
“When the SCA
conversation with a stateful session bean ends, for example as a result of a
call to ServiceReference.endSession(), the associated session object will be
removed by calling the EJB Home remove() method.”
Shouldn’t this really be a call
to the remove method on the EJBObject instead of the EJBHome? The
EJBHome interface has the following two remove
methods:
Public void
remove(javax.ejb.Handle handle);
Public void remove(Object
primaryKey);
The second remove method would
be illegal to invoke on a session bean home. The first remove method
implies a prior call to EJBObject.getHandle(). Further, the EJBLocalHome
interface only has the remove method that takes the primary
key.
PROPOSAL:
The remove method on the
EJBObject or EJBLocalObject interface should be used to remove the
EJB.