The I-desc is a nice idea, however you're going to get multiple hits on the same song, and don't we all hate that when we hear the same song?
The basic idea is very similar to something that I did with the thought of the day. I used to select the entire file put it into an array, and then base the random number based off of the time of the day with a modulo of a prime number close to the total number of records selected and use that to prime RND. Or maybe visa versa. It was a LONG time ago in a universe very very far away ;) This was similar to hashing in pick, and where I got the idea. Using the time, gave as close to a really random number as possible. Since I was only selecting one record a day, I would post a 1 to that record so it would not be selected the next time. When my select didn't return any records I would just clear all the 1's in the file and start over. Worked great! But but but, why not just get the free winamp www.winamp.com and just click on random??? Allen -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Dave Walker Sent: Tuesday, August 09, 2005 08:22 To: '[email protected]' Subject: RE: [U2] True random select lists? Yes, I would like a DIFFERENT 100 every time, otherwise the FIRST would have worked. The first thing I thought of was your basic routine, but I just wondered if I'd overlooked a SELECT option that I wasn't familiar with. I'm wondering which would be more efficient; the basic subroutine or Christophe's suggestion of an I descriptor? Thanks, -- Dave Walker 8..7 4(()) -:&:- -:&:- 8.74 .74(()) ((88.74 ..74 -:&:- ((88.74 * Peace > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] Behalf Of > Martin Phillips > Sent: Tuesday, August 09, 2005 10:46 AM > To: [email protected] > Subject: Re: [U2] True random select lists? > > > Hi Dave, > > > I've built a database of all my mp3s. Now I want to > generate a random list > > of 100 songs. s there a way to do that with a SELECT > statement, or must I > > muck around in basic to do so. > > This depends what you mean by "random". If you are happy to > agree that the > hashing process is effectively random from an external point > of view, simply > use > SELECT filename FIRST 100 > > However, you seem to want a different 100 records each time. > I cannot see > any way to do this aside from using a Basic program. Even > then, you will > have to start with a list of all the records and then extract > 100 items > randomly... > > OPEN 'filename' TO FVAR ELSE STOP > SELECT FVAR > READLIST LIST THEN > N = DCOUNT(LIST, @FM) > FOR I = 1 TO 100 > X = RND(N) + 1 > ID = LIST<X> > DEL LIST<X> > N -= 1 > NEXT I > END > > The above idea assumes that the list has over 100 items in it > > The basic problem here is that computers are supposed to be > repeatable. What > you want is random behaviour so you are unlikely to find it > in the standard > toolset. > > > Martin Phillips > Ladybridge Systems > 17b Coldstream Lane, Hardingstone, Northampton NN4 6DB > +44-(0)1604-709200 > ------- > u2-users mailing list > [email protected] > To unsubscribe please visit http://listserver.u2ug.org/ ------- u2-users mailing list [email protected] To unsubscribe please visit http://listserver.u2ug.org/ ------- u2-users mailing list [email protected] To unsubscribe please visit http://listserver.u2ug.org/
