The two versions do not have much more in common than that they were designed at the same time, use both LSM trees, and are meant for writing replicated databases in the sense tht you have full control over the database log, which then also acts as the replication log.
The C++ version is basically set of classes that implement LSM tree overlays and a log and means to keep both in sync. An example database on top of that (StringDB) is provided.
The created databases are not compatible.
... Felix