lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "M. Smit" <li...@Gangkast.com>
Subject Re: retrieve tokens
Date Wed, 22 Dec 2004 17:43:23 GMT
Erik Hatcher wrote:

>
> Highlighter does not mandate you store your text in the index.  It is 
> just a convenient way to do it.  You're free to pull the text from 
> anywhere and highlight it based on the query.
>
>> Furthermore, you are saying that the highlighter takes care of the 
>> corresponding field/words for me and pull up a context snippet? Ouch, 
>> why haven't I stumpled upon the sandbox....
>
>
> See a screenshot of it here:  http://www.lucenebook.com (going live 
> within a week!)
>
Oh bliss, Oh joy.. This is exactly what I'm looking for... I'll plunge 
in to it and let you know!

But for the other issue on 'store lucene' vs 'store db'. Does anyone can 
provide me with some field experience on size?
The system I'm developing will provide searching through some 2000 
pdf's, say some 200 pages each. I feed the plain text into Lucene on a 
Field.UnStored bases. I also store this plain text in the database for 
the sole purpose of presenting a context snippet.

If I were to use the Highlighter with a Field.Text, I will not use the 
database plain part altogether. But still I'm a little worried about 
speed/space issues. Or am I just seeing bears-on-the-road (Dutch saying, 
in plain English: making a fuzz about nothing)..

    Martijn

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


Mime
View raw message