lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Chris Lu" <chris...@gmail.com>
Subject Re: Lucene vs. Database indexing (RE: Indexing and Searching from within a single Document)
Date Tue, 08 Apr 2008 14:27:28 GMT
Agree with Nick that Jing's task doesn't really need Lucene if mostly
range search is needed. Database is good for range search.

But for his search on 'Select * from table where Name like
"%mymymy%"'. It's not a scalable solution for database. And using
Lucene makes a lot of sense.

The easiest way is to create a Lucene index, and apply range search on
the index.

-- 
Chris Lu
-------------------------
Instant Scalable Full-Text Search On Any Database/Application
site: http://www.dbsight.net
demo: http://search.dbsight.com
Lucene Database Search in 3 minutes:
http://wiki.dbsight.com/index.php?title=Create_Lucene_Database_Search_in_3_minutes
DBSight customer, a shopping comparison site, (anonymous per request)
got 2.6 Million Euro funding!


On Tue, Apr 8, 2008 at 6:56 AM, Duan, Nick <NDuan@mcdonaldbradley.com> wrote:
> I think this should be a new thread since it's a different problem.
>
>  Based on your description, I don't see any compelling reasons for you to
>  use Lucene just for indexing purposes, since you are not indexing text
>  docs as you indicated.  Claiming database of lacking performance is not
>  accurate and objective.  Your search queries have only the name field
>  and the two number fields in the where clause.  So you can perform
>  indexing on those two fields using database's own indexing mechanism,
>  i.e. the "create index ..." statement, if these fields are not already
>  declared as primary keys.  You will see dramatic performance improvement
>  of your search queries.
>
>  ND
>
>  -----Original Message-----
>  From: jing_gao@agilent.com [mailto:jing_gao@agilent.com]
>  Sent: Monday, April 07, 2008 1:19 PM
>  To: java-user@lucene.apache.org
>  Subject: RE: Indexing and Searching from within a single Document
>
>  Hi,
>
>  I have a similar question. Not heard back from anyone yet.
>
>
>  Dear Lucene experts,
>
>  I'm currently evaluating options for our search tool.
>
>  The need is:
>  I have millions of entries in database, each entry is in such format
>  (more or less)
>
>  ID      Name    Description     start (number)  stop(number)
>
>
>  Currently my application uses the database to do search, queries are in
>  the following format:
>
>  Select * from table where Name like "%mymymy%"
>
>  Select * from table where start >5 and stop <50000
>
>
>
>  I would like very much to use Lucene for such search, for the reason:
>  1. database performance is not ideal;
>  2. data is growing to be too big, I want to move to file system,
>  3. Currently everything is on server, user access through a web
>  application. I want to provide rich client tool, in which case I would
>  rather not to bother with database installations on client machine.
>  Database is my last option;
>  4. Lucene sounds very cool, I want to use a different technology than
>  database, which we are very familiar with already.
>
>
>
>  I read the book, played with the demo. My question is:
>  As you see, I'm not indexing or querying out documents, I'm interested
>  in one row of data. If I want to use Lucene, how should I do it? Do I
>  have to store my data as documents? Since I have millions and millions
>  of rows in database, if I store each row as a document, it'd be millions
>  of documents.
>
>
>  What do you suggest?
>
>  Thank you!
>  Jing
>
>  -----Original Message-----
>  From: syedfa [mailto:fayyazuddin@gmail.com]
>  Sent: Monday, April 07, 2008 10:13 AM
>  To: java-user@lucene.apache.org
>  Subject: Indexing and Searching from within a single Document
>
>
>  Dear Fellow Java/Lucene developers:
>
>  I am writing an application where a user is able to search for keywords
>  from
>  within a single book.  When the user conducts a search, he/she should
>  receive a set of results that show the sentence/phrase within the book
>  where
>  the keyword is found.  Unfortunately, all of the examples that I have
>  for
>  searching using Lucene discuss the concept of searching multiple
>  documents,
>  instead of within a single document.  I have written an application that
>  creates an index of this book, but I now want to search it.  In the
>  result
>  set, I would like the keyword(s) to be highlighted using Lucene's
>  HighLighter feature.  Once the user clicks on the hit from the result
>  set
>  list that they are looking for, the application should take them
>  directly
>  within to that section of the book where that keyword is found.  The
>  book
>  that I have indexed is in xml format.
>
>  My question is, how would I write an application that allows me to
>  search a
>  single document, and present a set of results to the user that list
>  portions
>  of text from the book that contains the user's keyword, instead of
>  presenting a list of document titles where that keyword is found?
>
>  Any help would be greatly appreciated.  Thanks to all who reply.
>
>  Sincerely;
>  Fayyaz
>  --
>  View this message in context:
>  http://www.nabble.com/Indexing-and-Searching-from-within-a-single-Docume
>  nt-tp16537558p16537558.html
>  Sent from the Lucene - Java Users mailing list archive at Nabble.com.
>
>
>  ---------------------------------------------------------------------
>  To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
>  For additional commands, e-mail: java-user-help@lucene.apache.org
>
>
>
>
>  ---------------------------------------------------------------------
>  To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
>  For additional commands, e-mail: java-user-help@lucene.apache.org
>
>
>  ---------------------------------------------------------------------
>  To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
>  For additional commands, e-mail: java-user-help@lucene.apache.org
>
>

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


Mime
View raw message