Subject Re: SEARCHING VERY LARGE FILE From Akshat Kapoor <akshat.kapoor@kapoorsons.in> Date Thu, 2 Sep 2021 16:40:30 +0530 Newsgroups dbase.getting-started Good Evening Mervyn, >> >>> This method works as a charm in medium/small files, it is very simple > ....... >>> I'll keep  on researching until I get a perfect method to get a good >>> incremental search for my large databases. >>> Once again thank you very much and have a wonderful day. >> >> Have you tried using a timer. >> I prefer waiting for 1 second before requery() > > Agostinho is using LIKE '%whatever%' to select records from over 1 > million records in a .dbf file.  This is a hard task for SQL and takes > long enough without adding a delay between sending individual characters > of the search string to the SQL engine. > > Agostinho, unless you change your approach and somehow limit the initial > number of records to search by introducing additional parameters, which > should preferably access indexed fields, into the select process you > will not find a software solution to speeding up your character by > character search of over 1 million records. > > Faster hardware for table access is the only viable solution to speed up > the existing search and even that can only go so far.  This would be > true even with a "proper" SQL server. It is just for hit and try. It will work with fast typers. Suppose the search string is abc Normal search takes 3 seconds. Total 9 seconds. With my method it will be 1+3 = 4 seconds provided abc are typed within 1 second. If it fails then no harm done. Just small coding was done. BUT if it works then cost of hardware upgrade will be saved. I have 2 forms. Both accessing the same table (around 3 k rows). One that is used many times daily has been optimised and works fine (No delay is observable) The other which is used few times a month has not been optimised using a timer and it is creating problems similar to Agostinho is facing. Regards Akshat