Mailing List Archive

a very basic question
How can a Field be indexed, but not stored? Isn't being indexed mean that
it is now in the index, and thus stored?

thanks,
rob


--
To unsubscribe, e-mail: <mailto:lucene-user-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:lucene-user-help@jakarta.apache.org>
Re: a very basic question [ In reply to ]
Rob,

> How can a Field be indexed, but not stored? Isn't being indexed mean that
> it is now in the index, and thus stored?

My assumption, which I have yet to check, was that there is a
difference between indexing and storing. I.e. indexing means making
notations in a table of "these words are in these locations".
Reconstructing the source document from that table would be
prohibitive. That, plus the ability to get a stored copy of the
document, plus the distinction in the API between indexing and
storing, lead me to suspect that there is a separate mechanism for
keeping a copy of the source document in or near the index.

I'll bet if you took a look at the source, you'd learn a lot. It
really is quite cleanly designed and coded. There's a bit of a
startup cost because a lot of the code has that lean, spidery look
that I often see with heavily-refactored projects, but you quickly get
past that.

Steven J. Owens
puff@darksleep.com


--
To unsubscribe, e-mail: <mailto:lucene-user-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:lucene-user-help@jakarta.apache.org>
Re: a very basic question [ In reply to ]
Thank you. That actually clears up a lot of my confusion. I've gone a bit
further in my code, to the point of doing searches into my index, and it
looks like everything is working quite beautifully.

I really didn't expect the indexes would be as small as they are, and so
assumed that it wasn't working properly.

One thing I've found that wasn't obvious from the documentation is that if
you want your indexed text to be analyzed, it's not enough to pass in the
analyzer to the indexer. I actually seem to have to index the text before
hand, and so I'm not quite sure what the analyzer is doing and why it
needs to be passed into the indexer.

thanks,
rob

http://www.robdecker.com/
http://www.planetside.com/

On Thu, 28 Mar 2002, Steven J. Owens wrote:

> Rob,
>
> > How can a Field be indexed, but not stored? Isn't being indexed mean that
> > it is now in the index, and thus stored?
>
> My assumption, which I have yet to check, was that there is a
> difference between indexing and storing. I.e. indexing means making
> notations in a table of "these words are in these locations".
> Reconstructing the source document from that table would be
> prohibitive. That, plus the ability to get a stored copy of the
> document, plus the distinction in the API between indexing and
> storing, lead me to suspect that there is a separate mechanism for
> keeping a copy of the source document in or near the index.
>
> I'll bet if you took a look at the source, you'd learn a lot. It
> really is quite cleanly designed and coded. There's a bit of a
> startup cost because a lot of the code has that lean, spidery look
> that I often see with heavily-refactored projects, but you quickly get
> past that.
>
> Steven J. Owens
> puff@darksleep.com
>
>
> --
> To unsubscribe, e-mail: <mailto:lucene-user-unsubscribe@jakarta.apache.org>
> For additional commands, e-mail: <mailto:lucene-user-help@jakarta.apache.org>
>


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