Maybe I'm way off here, but when you're looking for his numbered place in the record set, can't you just count the people before him? ID NAME STATE DATE 1 Joe Maine 2006-02-12 2 Jim Maine 2006-01-12 3 James Maine 2006-03-12 4 John Maine 2005-05-12 5 Jordan Maine 2006-02-11 Pseudo-sql below that is untested and probably has a typo in it: SELECT count(*) FROM employees WHERE STATE='Maine' AND DATE < (SELECT DATE FROM employees WHERE ID=1); You count the number of people that have a starting date before his and the same state. Wouldn't he be the next record after that? Depending on if you want the count from 0 or 1, change the < to <=. "James Tu" <jtu@xxxxxxxxxxxxx> wrote in message news:A3BF5223-6914-4416-BEC5-0E83DCFABD48@xxxxxxxxxxxxxxxx > I've cross posted this to the MySQL list... > > Here's my original post. > >> Is there some quick way to do the following in MySQL? (I know I can use >> PHP to search through the result set, but I wanted to see if there's a >> quick way using some sort of query) >> >> Let's say I know that Joe is from Maine. >> I want to do a query of all employees from Maine, ordered by hiring >> date, and figure out where Joe falls in that list. (i.e. which record >> number is he?) >> >> -James >> > > > Here's my new plan of attack... > > Right now I'm trying to use PHP to do a binary search on the result set > so I don't have to traverse the entire result set. > > I'm using PHP's mysql_data_seek() to move the pointer within the result > set and looking at the data to do the necessary comparisons. > > What do people think of this approach? > > -James -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
