I don't have any special knowledge of the OGNL implementation, but it seems that just wrapping a class to bolt on its accessors is not sufficient in the general case where the class is a descendant in a derivation hierarchy.
What isn't clear to me is why ognl would need to do any type of wrapper or subclass in the first place. The one place we definitely know this occurs is in Tapestry when an abstract page is subclassed to add the properties specified in the .page file. On 4/27/05, Patrick Casey <[EMAIL PROTECTED]> wrote: > Is ognl actually *subclassing* my object in order to bolt on its own > accessor methods? Or is it constructing (as I assumed, perhaps erroneously) > a proxy of some sort?). --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
