Okay. Thanks for the comprehensive information!
Regards
Felix
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4078773#4078773
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4078773
___
anonymous wrote : But if this is correct doesn't this problem apply to any case
where the scope of the datamodel is longer than that of the action bean
outjecting it? The same could happen with a stateless action bean and a
conversation scoped datamodel.
No because the only scope you can speci
I think I am finally getting you. IMHO it is acceptable that the backing list
would be null during the second call but than null would be outjected again and
the dataModel removed from its scope after the call which is probably a bit
confusing :-).
But if this is correct doesn't this problem ap
anonymous wrote : @DataModel is like @Out with additional logic to "convert"
the data from one type to another (e.g. List -> DataModel). However it has an
additional bit (reinjecting the value when using PAGE scope) - that which we
are discussing.
Add this explanation and the discussion result
So that when you do:
@Name("eventLister")
| public class EventListerAction {
|
| @DataModel(value="eventList", scope=ScopeType.PAGE)
| private List eventList;
|
| @DataModelSelection("eventList")
| private Integer selectedEvent;
|
| @Factory("eve
Maybe I am a bit thick here but I still don't get why a setter should be
needed. If the datamodel is page scoped I would expect it to go back into page
context (= Contexts.getPageContext().set("name", dataModel)) after
deserialization. What sense does it make to reinject it into the action bean
It's required if the DataModel has been outjected to PAGE scope (as the
component is event scoped but the datamodel isn't). I'll drastically reduce
the requirement for the setter by only calling when we are using PAGE scope :)
(n.b. previous to this change you would have got an IAE if you fall
Done.
http://jira.jboss.com/jira/browse/JBSEAM-1868
IMHO a setter should not even be required when susing @DataModelSelection. The
datamodel should be available from the context it was outjected to. Pre Seam
2.0 this was not required so this is a clear regression to me. On the other
hand addi
If you file a JIRA issue I can move the check for the setter to later on (you
will still need a setter for the DataModel if you are using @DataModelSelection
though)
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4078400#4078400
Reply to the post :
http://ww
bump
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4078271#4078271
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4078271
___
jboss-user mailing list
jboss-user@lists.jboss.org
10 matches
Mail list logo
