> I am not sure if the interface needs adaption? I'd say not > really, but > it would be cleaner...
I'd vote for not changing interfaces in order to save code existing somewhere outside Slide. Regards, Peter > -----Original Message----- > From: Oliver Zeigermann [mailto:[EMAIL PROTECTED] > Sent: Monday, November 10, 2003 13:59 > To: Slide Users Mailing List > Subject: Re: Security implementation: ACL draft-12 compliance > > > Nevermann, Dr., Peter wrote: > > > Could it be a problem with a particular store implementation? > > > > The interface provides 4 methods: > > > > - void grantPermission(Uri uri, NodePermission permission) > > - void revokePermission(Uri uri, NodePermission permission) > > - void revokePermissions(Uri uri) > > - Enumeration enumeratePermissions(Uri uri) > > > > The Javadoc of grantPermission() > [org.apache.slide.store.SecurityStore] does > > not explicitly state it ... but at least in our Tamino-based store > > implementation, the new ACE is *appended* to the existing > ACL. I suppose > > that Oliver's TX store does it in the same way. And > enumeratePermissions(), > > of course, should return the ACEs in the defined ordering. > > That's correct, but this is an implementation detail, not something > required by the interface, isn't it? Certainly, the RDBMS > implementations need to store the order of entries, which > they currently > do not do. > > I am not sure if the interface needs adaption? I'd say not > really, but > it would be cleaner... > > Oliver > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] >
