Hi,

On Tuesday, Jun 10, 2003, at 22:03 Europe/Rome, Otis Gospodnetic wrote:

> [...]
> On a side, somebody has already written a tool to make a Lucene index
> out of DB tables.  Why re-invent the wheel?

we have build our own application to index the content of a DB.
Our application is not "general purpose", but is exactly wrapped on our 
current needs.
We have been able to replace Intermedia indexes with excellent results, 
both in indexing an searching time.

At the moment we have almost 3 milions record indexed with about 40 
thousand we documents inserted every day.

If anybody is interested in a deeper description on how we did achieve 
these results, just let me know.
We used a mix of support table / trigger on the DB side so our solution 
won't scale well a general purpose DB to Lucene indexing engine, but 
someone could get some interesting ideas.

As the full description will be quite lenghty, I don't want to write it 
without anybody interested in reading it! ;-]
Otherwise, I would be very pleased to give something back to the Lucene 
comunity, as Lucene did help us very much in settling down our 
search/indexing problems.

Regards,

Giulio Cesare Solaroli

PS: no problem also on sharing our code.


---------------------------------------------------------------------
To unsubscribe, e-mail: lucene-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: lucene-dev-help@jakarta.apache.org