On Jan 19, 2009, at 3:34 PM, Karl Wettin wrote:
How do we feel about adding dependecies to BerkeleyDB in the core? I
know we talked about it before, but as we are talking about moving
to Maven we won't be bundling any libraries. Does that help?
Yeah, I think it's fine by me.
I use it in the heirarchial cluster and in the LUCENE-626
"reinforcement learning" spell checker. Both of these could be
popped in the trunk if BDB was not a problem.
One alternative is that I made some abstract implementation and then
a contrib-module with BDB bindings.
That usually is the best way to do these things anyway.
