Op 1/13/10 12:43 AM, Robert Cummings schreef: > > dealtek@xxxxxxxxx wrote: >> On Jan 12, 2010, at 1:57 PM, Ashley Sheridan wrote: >> >>> Depends on how you're creating running the query. You could do >>> something like: >>> >>> echo mysql_result($result, 1, 'fieldname'); >>> >>> Where $result is your result object and 1 is a 0 indexed array, so >>> would be the second result. >> >> Thanks Ryan, Ashley & Kim for the good techniques... >> >> - in my case I was trying to pull a random record from a query of >> Table1 - then do a 2nd query from 1st so >> >> mysql_result worked fine for my needs like: >> >> >> ... do query 1... 'cur' - SELECT id FROM myTable1 >> >> $ran = rand(0, $totalRows_cur - 1); // pick random rec row within >> total count >> >> $pick1 = mysql_result($cur, $ran); // get the ID from the choice - >> (like id=252) >> >> ... do query 2 // where relatedID = $pick1 of myTable2 > > Put your random logic into the query: > > SELECT > something > FROM > somewhere > WHERE > condition > ORDER BY > RAND() > LIMIT > 1 please read the following to understand the consequences and possible performance problems with this approach: http://www.titov.net/2005/09/21/do-not-use-order-by-rand-or-how-to-get-random-rows-from-table/ personally I go with the 2 query approach ... or even to use a specific field to store a generated random value (with suitable index) for each record, the stored values can be regenerated in an out-of-band process (e.g. cron job) ... such a setup would suffice in situations where frontend response speed is of primary concern and it's acceptable to return the same 'random' record over a given period (i.e. the time between random value regeneration) > Cheers, > Rob. -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
