Looking at the code I see the problem is org.apache.wicket.model.CompoundPropertyModel.setChainedModel(IModel<?>)
On Wed, Sep 15, 2010 at 4:27 PM, Martin Grigorov <[email protected]>wrote: > Don't hesitate to provide patches ;-) > > > On Wed, Sep 15, 2010 at 4:06 PM, Ernesto Reinaldo Barreiro < > [email protected]> wrote: > >> Shall I provide a patch for 1.5 and attach it to a JIRA ticket? >> Probably this change cannot be applied to 1.4.x (as it might break >> existing apps) >> >> Ernesto >> >> On Wed, Sep 15, 2010 at 4:02 PM, Martin Grigorov <[email protected]> >> wrote: >> > Looks good to me. >> > I've never seen that this ctor looks this way and I don't remember >> > discussion related to that. >> > >> > On Wed, Sep 15, 2010 at 3:22 PM, Ernesto Reinaldo Barreiro < >> > [email protected]> wrote: >> > >> >> My apologies if this has been discussed before... >> >> >> >> Right now the constructor of CompoundPropertyModel is as follows: >> >> >> >> /** >> >> * Constructor >> >> * >> >> * @param object >> >> * The model object, which may or may not implement >> >> IModel >> >> */ >> >> public CompoundPropertyModel(final Object object) >> >> { >> >> target = object; >> >> } >> >> >> >> which allows users to do something like: >> >> >> >> new CompoundPropertyModel<XXX>(new YYY()); >> >> >> >> without generating a compiler error. Would it make sense to replace >> >> this constructor by two constructors... >> >> >> >> /** >> >> * Constructor >> >> * >> >> * @param object >> >> * The model object >> >> */ >> >> public CompoundPropertyModel(final T object) >> >> { >> >> target = object; >> >> } >> >> >> >> /** >> >> * Constructor >> >> * >> >> * @param object >> >> * an instance of IModel<T> >> >> */ >> >> public CompoundPropertyModel(final IModel<T> object) >> >> { >> >> target = object; >> >> } >> >> >> >> at least on 1.5? >> >> >> >> Regards, >> >> >> >> Ernesto >> >> >> > >> > >
