On 10/23/01, David Burry wrote:
> In fact I didn't even compress the inverted index at all, used
> Berkley DB (B+tree) instead which is quite wasteful of space (makes
> it take about 25 megs per version) but it sure is
> ***blazing***fast*** as a result!!!
I've got what may be an "inverted index" for the same purpose, using
the Berkeley B+tree database as well. Maybe I just don't understand
what an inverted index is, but mine ends up be about 4.3mb per version
for the database. Keys are the words themselves (no punctuation
except what may exist within a word), in lower case. Values are
packed integer arrays, each integer being the verse index where the
word is found. If you're recording more than that, I'd like to know
what your database looks like.
Keep in mind I've only taken a handful of programming courses, so I
might be a little hazy on some terms.
Jesse
--
If we confess our sins, He is faithful and just to forgive us our sins
and to cleanse us from all unrighteousness. If we say that we have
not sinned, we make Him a liar, and His word is not in us.
http://www.grace-els.org
