|
From: | paul POULAIN |
Subject: | [Koha-devel] killer feature ? |
Date: | Fri Oct 17 02:29:13 2003 |
User-agent: | Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624 |
Hi, I spoke with Rachel yesterday, and we spoke of something the could be a "killer feature" in Koha. In google, when you type a word wrong and get no or few answers, google says "you could also try somethingAlmostSame" the SomethingAlmostSame being in the DB. In Koha, we could add this feature really easily (1 SQL query get the answer :-) ). I did some testing with the following SQL (on Dombes Abbey DB) : SELECT count( * ) AS total, word FROM marc_word WHERE sndx_word = soundex( "Raner" ) AND word <> "Raner" GROUP BY word ORDER BY total DESC It answers : total word 83 Rahner 1 rumeur rahner being the right author name :-) another test (with "Jésu", instead of Jésus) : 245 Jésus 182 Jacques 178 JEA 18 José 15 JOU 12 Joie Note : 1- the SQL query is indexed, so very fast. 2- the major problem being to add this possibility when the search returns less than X results (important GUI coding imho) What do you think of this idea ? Anyone volunteer ? -- Paul POULAIN Consultant indépendant en logiciels libres responsable francophone de koha (SIGB libre http://www.koha-fr.org) |
[Prev in Thread] | Current Thread | [Next in Thread] |