On Tue, Mar 6, 2012 at 3:15 PM, Marcel Reutegger <[email protected]> wrote: >> How can we guarantee ids are unique *and* sequential in a >> cluster without (too much) overhead? > > use the path as id instead of generating some arbitrary id? > > after all, most methods of the JCR API operate on a path > and not an id.
And with moves the id stays the same? What happens in this scenario: /foo/bar/lux is moved to /foo/bar/flux and you add a new node 'lux' at /foo/bar If the move did not change the id, the addition of 'lux' later on cannot have the id of its path. Regards Ard > > regards > marcel -- Amsterdam - Oosteinde 11, 1017 WT Amsterdam Boston - 1 Broadway, Cambridge, MA 02142 US +1 877 414 4776 (toll free) Europe +31(0)20 522 4466 www.onehippo.com
