Re: dbms_random question

OK, thank you for the explanation!

My additional thanks also to Ed Prochak and Hans Forbich. I hope, that as soon as I get more experiences, I'll be able to help you :-)

Pavel

"Jim Kennedy" <kennedy-down_with_spammers_at_attbi.com> píše v diskusním příspěvku news:NT5za.944066$F1.117195_at_sccrnsc04...
> No, I am correct. Random number generators generate numbers that are with
> replacement.(they will repeat) In the case of THIS random number
generator
> you need to give it a seed to start its sequence off. If you do not give
it
> a seed then it always returns the same number. It is quite common to have
a
> random number generator that will return a specific sequence once it has a
> specific seed. This is a feature. If you were doing QA then you might
want
> a case that repeats its sequence of random numbers.
>
> You need to understand that even with your seed it will repeat at some
> point. I've run into this misconception about random numbers quite often
> especially if people try to use them as a primary key. They are surprised
> when they get an error because the number has repeated.
>
> Did you read the docs on this package?
>
> Jim
>
> --
> Replace part of the email address: kennedy-down_with_spammers_at_attbi.com
> with family. Remove the negative part, keep the minus sign. You can
figure
> it out.
> "Pavel Vetesnik" <Pavel.Vet_at_volny.cz> wrote in message
> news:baioad$2hp3$1_at_ns.felk.cvut.cz...
> > Hi,
> >
> > thank you for your posting, but I guess, you are wrong - otherwise I was
> > extremly lucky and got ten times the same number.
> > After calling the
> > ------------------
> > dbms_random.initialize(5);
> > dbms_random.random;
> > ------------------
> > I always got the 1034085445.
> >
> > But it is all right now - the
> > ------------------
> > dbms_random.initialize(TO_NUMBER(TO_CHAR(SYSDATE, 'SSSSS')));
> > ------------------
> > cured the problem. But it is question why use the dbms_random package
and
> > not call the TO_NUMBER(TO_CHAR(SYSDATE, 'SSSSS')) directly. It gives
> > "random-enough" numbers :-)
> >
> > Pavel
> >
> > "Jim Kennedy" <kennedy-down_with_spammers_at_attbi.com> píše v diskusním
> > příspěvku news:_25za.695898$Zo.149563_at_sccrnsc03...
> > > Random does NOT mean without repeating. If you have a jar with 10
> marbles
> > > and each marble has a different number on it. The random number
> generator
> > > is like choosing a marble and then putting the marble back in the jar
> and
> > > choosing again. This is called choosing with replacement. You can
get
> > the
> > > same number again. If you are looking for a random number generator
> that
> > > draws a new marble each time and the does not replace the marble back
> into
> > > the jar then you are going to have to keep track of all the marbles
> drawn
> > > and ignore those that came up before.
> > > Jim
> > >
> > > --
> > > Replace part of the email address:
kennedy-down_with_spammers_at_attbi.com
> > > with family. Remove the negative part, keep the minus sign. You can
> > figure
> > > it out.
> > > "Pavel Vetesnik" <Pavel.Vet_at_volny.cz> wrote in message
> > > news:baigpb$2e53$1_at_ns.felk.cvut.cz...
> > > > Hello,
> > > >
> > > > I want to generate a random number (to create an unique password). I
> > > > thought, the dbms_random package was created for this purpouse.
> > > >
> > > > I use such code:
> > > > ============================
> > > > nMagic NUMBER(11);
> > > >
> > > > dbms_random.initialize(5);
> > > > nMagic:=dbms_random.random;
> > > > dbms_random.terminate;
> > > > ============================
> > > >
> > > > But I always get the same number (1034085445 in this case)!
> > > > Giving different 'seed' parameter (5 above), I can get different
> > numbers,
> > > > but the same 'seed' means the same number.
> > > > So what that means? To create a random number I must pass a random
> > number
> > > to
> > > > the generator?!
> > > >
> > > > Please can you help me with this?
> > > > Thank you in advance,
> > > > Pavel
> > > >
> > > >
> > >
> > >
> >
> >
>
>
Received on Fri May 23 2003 - 03:00:29 CDT