You are right! This will solve the problem. But for some reason I expect Wicket to do this for me - the model is IComponentInheritedModel and it has an "owner", the component that owns the model. So it should be feasible to calculate the relative path between them. I'll investigate whether this is possible.
Martin Grigorov Wicket Training and Consulting https://twitter.com/mtgrigorov On Thu, Aug 7, 2014 at 3:40 PM, John Sarman <[email protected]> wrote: > Martin, > > I think you need the address wicket:id changed to person.address and the > street1 to person.address.street1 . > > WebMarkupContainer addressContainer = new > WebMarkupContainer("person.address"); > > addressContainer.add(new Label("person.address.street1")); > > John Sarman > > > On Thu, Aug 7, 2014 at 9:30 AM, Martin Grigorov <[email protected]> > wrote: > > > Hi, > > > > Debugging https://issues.apache.org/jira/browse/WICKET-4941 I found out > > that CompoundPropertyModel works only shallow component hierarchy (one > > level deep!). > > Is this by design ? > > > > I've added a unit test that shows the problem: > > > > > https://git-wip-us.apache.org/repos/asf?p=wicket.git;a=commitdiff;h=7042f885;hp=98404f5541fee8064ff01dee7ff7f4209669245c > > > > If you have a container for Person with two children - one for name > (Label) > > and another for Address (a container). The Address container have its own > > children, like street (Label). > > With the current impl of CompoundPropertyModel (and > Component#initModel()) > > Wicket tries to find "street" in Person, while it should try > > "address.street" in Person. > > > > Anyone with more experience with IComponentInheritedModel could help > here ? > > > > > > Martin Grigorov > > Wicket Training and Consulting > > https://twitter.com/mtgrigorov > > >
