+1
On 8/29/07, Igor Vaynberg <[EMAIL PROTECTED]> wrote: > > +1 > > -igor > > > On 8/29/07, Igor Vaynberg <[EMAIL PROTECTED]> wrote: > > > > seems this is confusing a lot of users and it doesnt work the greatest, > so > > here is a vote > > > > so the vote is to deprecate the constructor variants that do not take a > > markup provider as an arg. > > > > -igor > > > > > > On 8/29/07, Oli Evans <[EMAIL PROTECTED]> wrote: > > > > > > I just lost a few hours on a problem where an ajax operation on a > Panel > > > nested in a Fragment was failing as my nested panel could not find its > > > associated markup. It seems this was because a Fragment will only > > > reliably find its markup if you explicitly provide it with the > Component > > > > > > that contains its markup in the Fragment constructor. > > > > > > If you don't, Fragment.getAssociatedMarkupStream() will find its > > > markupProvider field is null, and so have to take a guess at where its > > > markup might be. If your fragment usage happens to be as a direct > child > > > of the component that defines the fragment markup then it will all > work > > > out, but if you have nested it in something like a ListView then it > will > > > fail. > > > > > > ...So watch out kids, if you use fragments, use a constructor that > tells > > > > > > it where its markup is. Would you like to have to guess who your real > > > parents are? I think not. > > > > > > Oli > > > > > > --------------------------------------------------------------------- > > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > > > >
