> If it is not "in the works", maybe I will take another whack at it (the
> manual PI pruning thing that is
Be aware that pruning with the DTM will have some new challenges...
especially how to reuse handle space. The DTM is stored in a bunch of
arrays, with the lower 20 bits of the Node handle referring to what is
basically it's row index. Pruning with the DTM will basically mean that
you want to start reusing the rows that belong to a subtree, since you
can't really re-pack the DTM arrays. Because the node handle also
represents a document order number that is used for one of the API calls,
this can get tricky.
While the DTM made life in some ways much easier, it complicated things in
other ways. It is also a design challange for the SQL extension, though it
also presents some good oppurtunities.
-scott
