lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Catalin Teodorescu" <cteodore...@MusicNet.com>
Subject RE: name search from Oracle to Lucene
Date Tue, 13 May 2003 16:13:26 GMT
If you have a table with a column NAME and an index on this columns then:
1. select NAME from T where NAME like "a%" will USE the index
2. select NAME from T where NAME like "%a" will NOT USE the index
3. select NAME from T where NAME like "%a%" will NOT USE the index
4. select NAME from T where NAME like upper("a%") will NOT USE the index
   for this you have to create a function index with function being upper(name) (see oracle
documentation)
   another alternative will be to have a different column NAME_UPPER and use this one for
searches.

Now depending on your need (complex search criteria: partial matches, ANDs, ORs ... a straight
index will not help).
Another alternative is to use Oracle Intermedia.

You can also try lucene. It is great.

We have such tables/searches and we are indexing them with lucene and the searches are very
fast.
Basically we have an app doing the indexing and a lot of other servers doing the searches.

Catalin

-----Original Message-----
From: david chan [mailto:david_tomcat@yahoo.com]
Sent: Tuesday, May 13, 2003 08:57 AM
To: Lucene Users List
Subject: name search from Oracle to Lucene


Hi,
 I have a table with more than 1 million records, I
need to some search on these three fields: firstname,
lastname, nickname
 Now when I try to search these names from Oracle with
some keyword "like%", or function for "upper" to
ignore case compare, these kind of query become very
slow even these name fields are indexed. One reason is
because these function "like", "upper" not using the
index.
 I wonder it can be done better in Lucene, anyone can
give me some idea how name search (partial name
search, firstname search, lastname search, nickname
search, or combination search) can be done?
 I am thinking to dump this big oracle table to a
plain text file, then read in and index by using
Lucene. But now sure how to arrange these fields to
get a good/fast result.

Thanks.
David

__________________________________
Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo.
http://search.yahoo.com

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


---------------------------------------------------------------------
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