[BNM] Suggest in search - way of doing with MySQL?

[wayne at codingvista.com]

And the wiki: http://en.wikipedia.org/wiki/Lucene

:)

-----Original Message-----
From: bnmlist-bounces at brightonnewmedia.org
[mailto:bnmlist-bounces at brightonnewmedia.org] On Behalf Of
wayne at codingvista.com
Sent: 02 October 2008 10:55
To: 'Brighton New Media'
Subject: Re: [BNM] Suggest in search - way of doing with MySQL?

The levenshtein algorithm is only part of the solution. I tried to write
this as a .NET search tool for a project and it ended up getting more and
more complex as it's short fallings became apparent.

The Lucene fuzzy search implementation is a bit of a life saver for this
kind of thing.

PHP: http://devzone.zend.com/node/view/id/91
Java: http://lucene.apache.org/java/docs/queryparsersyntax.html
.NET: http://incubator.apache.org/lucene.net/

W://

-----Original Message-----
From: bnmlist-bounces at brightonnewmedia.org
[mailto:bnmlist-bounces at brightonnewmedia.org] On Behalf Of James Lawrie
Sent: 02 October 2008 10:49
To: Brighton New Media
Subject: Re: [BNM] Suggest in search - way of doing with MySQL?

I'm not sure how *easy* it will be but you want to implement a
levenshtein distance algorithm. PHP has it built in:
http://uk.php.net/manual/en/function.levenshtein.php

I found an implementation here for mysql too:
http://codejanitor.com/wp/2007/02/10/levenshtein-distance-as-a-mysql-stored-
function/

It's not a complete answer, but hopefully it will save you a bit of time.
-- 

BNM Subscribe/Unsubscribe:
http://www.brightonnewmedia.org/options/bnmlist

BNM powered by Wessex Networks:
http://www.wessexnetworks.com

-- 

BNM Subscribe/Unsubscribe:
http://www.brightonnewmedia.org/options/bnmlist

BNM powered by Wessex Networks:
http://www.wessexnetworks.com