Ah I see, your addTerm() is heading towards the NestedAssertion concept ... however, I need to take another look at the interface since I'm not that convinced that Nested behaviour is handled adequetly ... more on this anon.
Regards, Brian. On 5/5/06, Sanka Samaranayake <[EMAIL PROTECTED]> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello Everyone, Dmitry Goldenberg wrote: > Hello, > > Question on the addTerm method on PrimitiveAssertion. Why is this method there? Isn't it the point of PrimitiveAssertion that you can *not* add to it, i.e. it is not composite but finite? The name PrimitiveAssertion does not necessarily mean that it is a leaf node. It could well contain another assertions as its child assertions. The main difference of a PrimitiveAssertion is that , it is domain specific. For instance a security assertion is considered as a primitive assertion. If you have a closer look at the WS Security Policy specification, you will realize that there are few security assertions which contains some other child assertions. Further more, those child assertions could be organized using standard policy operators like Policy, ExactlyOne and All which allows the same primitive assertions to have multiple configurations. Hence to allow such behaviour there is the need to have the method addTerm(..) in the API of the PrimitiveAssertion. Best, Sanka > > It seems to me that "addTerm" is better suited to be on the CompositeAssertion interface. Furthermore, should it not be called "addAssertion", to be more in line with the spec? The spec never uses the word "term." What is a term anyway? > > Thanks for any insight. > - Dmitry > -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.1 (GNU/Linux) iD8DBQFEW5Y+/Hd0ETKdgNIRAkkLAJ9U/sjmPxdbp/wBxy4sc79ZIVUfYgCeK9lR Pu/5nZgE5GU0rtU/PXVNovQ= =0Vij -----END PGP SIGNATURE-----
