On Tuesday 22 January 2002 12:32 pm, Cees de Groot wrote:
> Scott A Crosby <crosby at qwes.math.cmu.edu> said:
> >Hack'er one up yourself? Reverse indexes aren't all that slow or large.
> >Roughly comparable to the size of the origional data.
>
> I was thinking about just that. The only thing I am mainly concerned about
> was the best way to store the stuff on disk.

I'd use BerkeleyDB; it's portable, fast, and should be easy enough to make a 
plugin for.

There's already an FFI implementation for it.

-- 
Ned Konz
currently: Stanwood, WA
email:     ned at bike-nomad.com
homepage:  http://bike-nomad.com