Thanks, Stuart!

Yeah, I just noticed this part myself just before I saw your post.

Thank you!

Cheers,
=dml



On Wed, 2008-11-12 at 12:02 +0800, Stuart McCulloch wrote:
> 2008/11/12 David Leangen <[EMAIL PROTECTED]>
>         
>         I am using the KF impl. I thought I understood the model, but
>         maybe I'm
>         wrong, so I wanted to check here first.
>         
>         I have a User who implies a basic membership of both GroupA
>         and GroupB
>         (not a direct member, but membership is implied).
>         (Tested and both evaluate to hasRole=true)
>         
>         Both GroupA and GroupB and required members of TargetGroup.
>         (Also checked, and they really are members)
>         
>         So, if both GroupA and GroupB are required members of
>         TargetGroup, and I
>         can verify that:
>         
>          userAuth.hasRole( groupA.id ) == true
>         
>             &&
>         
>          userAuth.hasRole( groupB.id ) == true
>         
>         Should this then not evaluate to true??
>         
>          userAuth.hasRole( targetGroup.id );
>         
>         
>         It evaluates to false in my case...
>         
>         Am I misunderstanding something, or do I need to signal a bug
>         for the KF
>         code?
> 
> from the compendium spec, p158:
> 
>   "For example, an initiator Role object X implies an action Group
> object A if:
>     • X implies at least one of A's basic members, and
>     • X implies all of A's required members."
> 
> so does targetGroup have any basic members?
> 
> (btw, you can add Role.USER_ANYONE as a basic member to match anyone)
>  
>         Thanks!
>         =dml
>         _______________________________________________
>         OSGi Developer Mail List
>         [email protected]
>         https://mail.osgi.org/mailman/listinfo/osgi-dev
> 
> -- 
> Cheers, Stuart

_______________________________________________
OSGi Developer Mail List
[email protected]
https://mail.osgi.org/mailman/listinfo/osgi-dev

Reply via email to