Hi, On Mon, Jun 23, 2014 at 4:23 PM, Thomas Mueller <[email protected]> wrote: > Sorry, sure, the condition is verified again. But this might be an > in-memory operation. The index may return the property value for each > entry as part of running the query (QueryIndex - Cursor - IndexRow). I > think the index implementations don't do that currently, so in reality the > node is always loaded with the current version of Oak, that's true. > Javadoc of Cursor.next(): "The index should return a row with those > properties that are stored in the index itself, so that the query engine > doesn't have to load the whole row / node unnecessarily."
Ah, I see where you're going! It would indeed be nice if the index could also provide the data to back more than the virtual properties (jcr:score, etc.) it now does. Then a query like "SELECT a, b, c FROM ..." with an index that covers a, b and c could avoid the (1 + ...) term from OAK-1910. But getting to that point may be a bit tricky, especially because of access control. BR, Jukka Zitting
