Hey All,
So I'm still going through revisions of my project trying to get an idea of the best way to handle everything. Right now I am at a loss as to how to set up my model. I was very happy with my model setup when using only Flash Remoting, but we've moved over to FDS, but my model doesn't seem to make sense anymore. Currently, I have my model structured so that each "node" in the model is a subclass of a specific bean class, and also defines all of the relationships with the other data in the model (child data, etc.). This works really well, the code is concise, easy to read and manage, and doesn't at all seem like it's going to work with FDS (though I haven't yet tried). I see the issue that when I create my [Managed] bean, and subclass that into one of my "Model" classes, FDS won't be able to manage that anymore. Is this the case? Can I leave it [Managed], and just use the [Managed] bean's ArrayCollections independently? When FDS syncs the data, does it create a new object or simply load data into the existing one? Is it going to try to send all of the data in the object (including all data in the arraycollections? Ew.) to the server during syncs? Originally I had my model set up so that I had a number of Model classes that would contain an ArrayCollection of all of the data, and "beside that" a reference to another Model object that has references to the "current" object in that ArrayCollection and more ArrayCollections for child data. But that was long-winded, hard to maintain and hard to read. Lastly, how have you structured your Model? Do you have a best practice that you're working with that you prefer? This is sort of a spin-off of another topic discussing whether to store DataServices in your service locator or just instantiate them when you need them. All thoughts are appreciated. Jay Proulx Application Developer Internet Delivery Services - eVision Team [EMAIL PROTECTED] (613) 783-6711
