> 
> > how would you build the path of a node accessed by uuid?

> Jukka Zitting wrote:
> Instead of the current node id's, the search index would need 
> to contain the paths of the indexed nodes. Thus the 
> //[EMAIL PROTECTED]:uuid='...'] query (or its equivalent) executed by 
> getNodeByUUID() would return the path that can be used to 
> access the node.
> 

Doesn't this imply that moving a node would need re-indexing all its
child nodes because of changed paths, making a move a very expensive
operation? 

Regards Ard

> 
> BR,
> 
> Jukka Zitting
> 

Reply via email to