Marc Wick <[EMAIL PROTECTED]> writes: > be careful with random() as it will be difficult to achieve the same > semantic, if more than one row is updated. Just replacing it with a > single random value will break applications relying on every row being > assigned an other random value.
Errr... since when random values are guaranteed to be different to each other? "5,5,5,5,5" is as random as "1,2,3,4,5" or "6,5,2,4,5" or anything else... I think that only very specific applications making an intensive and statistical use of random() would break. Do you have examples? Lotteries? _______________________________________________ Sequoia mailing list Sequoia@lists.forge.continuent.org https://forge.continuent.org/mailman/listinfo/sequoia
