Hi Rich,

I don't think that I have a great answer for you on this but I have a few
questions/thoughts.

It sounds as though you want the PageItem's position to initially be set
from the VO's x/y, then you want the VO's x/y to reflect the PageItem's x/y
as it is dragged.

Why don't you *not* have the x/y in the VO bindable? Instead, just have them
bound to the PageItem's x/y.

There are at least two possible objections to this:

a) "How will the PageItem's x/y initially get set to the VO's x/y?" Well,
you'd have to figure out some other mechanism.

b) "But part of the definition of a VO is that all properties are bindable!"
Well, rules are made to be broken.  :)

But I'm not sure that this is a good idea. Bi-directional binding doesn't
sound bad to me. You say it's inefficient, but is it impacting performance?
If not, I wouldn't worry about it.

Douglas

On undefined, Rich Rodecker <[EMAIL PROTECTED]> wrote:

>   Hello,
> I'm building a page layout-type app using cairngorm. I have an
> arraycollection of value objects in the model that stores the data for each
> item on the page.  Every item on the page is a custom component (a PageItem)
> that is bound to one of the VOs as a dataprovider.  The VO contains the
> entire state of the PageItem, x,y,rotation, content, etc).
>
> each PageItem is draggable.  When I drag the PageItem to a new position, i
> need to have the VO updated as well.  Currently I am just updating the vo
> from within the component when the mouse is released, but since it's bound,
> that triggers the property change event on the VO, which doesn't really do
> anything but is pretty inefficient.
>
> I feel like there's gotta be some way to just proxy the PageItem's values
> like x and y to the VO directly and I'm just missing it.
>
> In the past I have used Momentos to save and restore state, but I'm trying
> to achieve a greater separation of the view and the model, so that when I go
> to save the data, I just grab the collection of VO's from the model and save
> that, rather than query the view for it's state.
>
> Anyone have any pointers?
>
>
>
> 
>

Reply via email to