> More comments on the documentation inside node-reference.xq: > > 1. I think the definition of "immutable" should say: > > i.e. a node identifier does not change during the node lifetime, and cannot be > reused for another node after the original node gets deleted. > > instead of just > > i.e. a node identifier does not change during the node lifetime >
Done. > 2. I don't like this sentence: > > "Identifiers are also unique, in that, two nodes with different identities > will never have identical identifiers.". > > Identifiers are actually an implementation of the "node identity" concept > defined by the w3c specs. So the above sentence is a tautology. I would say > the following: > > "Identifiers are also unique, in that, two different nodes will never have the > same identifier.". Done. > > 3. When we say "any node", I don't think it is necessary to add "either > temporary or stored in a collection". What is a "temporary" node after all? > There is no definition for "temporary" or "persistent" nodes anywhere. > Done. > 4. Should we say something about text nodes, eg, that their identifier is not > really immutable and it can be used to dereference the node only during the > same snapshot in which the identifier was retrieved? This refers to the other module, the one with the structural uris. Identifiers/references work the same for every node. Structural uris cannot be dereferenced (our choice), so it is not a problem. We never say however that the structural relationship of a non-text node remains the same in the node lifetime. Maybe we should. What do you think? -- https://code.launchpad.net/~zorba-coders/zorba/identifiers/+merge/78383 Your team Zorba Coders is subscribed to branch lp:zorba. -- Mailing list: https://launchpad.net/~zorba-coders Post to : [email protected] Unsubscribe : https://launchpad.net/~zorba-coders More help : https://help.launchpad.net/ListHelp
