When running the Recover tool on very large (>6G) databases, some statistics were reported with negative numbers.

Queries "group by" that contain a subquery with an aggregate function returned the wrong result in some cases.

New feature from Davide Cavestro - allow using custom Java object serialization engines on a per-DB basis.

when the search key was a different cardinal type from the table index key.
e.g. where the one was INT and the other was LONG

make NonUniqueHashIndex directly subclass BaseIndex. There is no point in sub-classing HashIndex if it is going to override all of the logic anyhow.

MVStore: use a separate write version for every map, so that the meta map version matches the version of the other maps.