On 18.07.2012, at 16:17, Jukka Zitting wrote: > In fact, the more I think of the specific case of the proposed > ":childOrder" property and orderability of nodes, the more it seems > that you are right - it is better to hide these details below the Oak > API and explicitly extend the API with an orderBefore() method like > the one in JCR.
FWIW: It would be nice to have a layer in oak that can be used for backups (or "batch export/imports", like SQL dumps) that contains all data and can be reimported without any (high-level) validation. In JCR/Jackrabbit so far this wasn't possible, e.g. versions could not be re-imported as those protected items could not be written via the JCR API without any validation (and there was no other API in Jackrabbit to do such a thing, except on the level of persistence manager implementation). ":childOrder" sounds like a similar internal or protected property. Having to call an orderBefore() instead of "just" importing a simple tree structure of (ordered) properties makes things more difficult for a backup approach. Regardless what API layer it will be, but IMHO there should be one by design that would allow such batch imports. Cheers, Alex
