Re: Question on Case-insensitive queries

From: Chris Jack <chrisj_at_pro-data.demon.co.uk>
Date: Mon, 12 Jul 1993 23:00:13 +0000
Message-ID: <742518013snz_at_pro-data.demon.co.uk>


In article <CA20wv.HIA_at_isst.fhg.de> umatz_at_isst.fhg.de writes:

>Sudhakar Bhagavatula (sab3_at_Ra.MsState.Edu) wrote:
>: Hi,
>: Could someone please tell the method to perform Case-insensitive
>: queries. Please send me an e-mail to:
>: sudhakar_at_cs.msstate.edu
>
>Hi,
>i want to know the answer to this question, too.
>Can you please post a summary or mail me direct.
>
I previously mailed information about this to Sudhakar, but as there seems to be interest, I will post it as well.

I personally implement strip searching. This is case insensitive, blank and punctuation stripped searching: ie 'I.., b _at_!m' matches with 'IBM'.

I simply have a routine to do the strip and then I store the non-stripped and stripped versions of the string in separate same sized fields on the database. They are both indexed, the former uniquely, the latter non-uniquely.

When the user does certain searches, it matches against the stripped version and returns any unstripped versions. In other circumstances, the search is done against the unstripped version.

-- 
Chris Jack
Received on Tue Jul 13 1993 - 01:00:13 CEST

Original text of this message