perl -MList::Util=shuffle -e 'print "@{[shuffle(1..10)]}"'

Re^2: How to not print the same random number

shotgunefx

created: 2006-03-04 06:01:55

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

While fine for small ranges, won't that be a hog on larger number ranges?
update I understand that the poster is only looking for 1-10, just seemed the better way to teach. Opinions?

-Lee
"To be civilized is to deny one's nature."

Re^3: How to not print the same random number

borisz

created: 2006-03-04 06:05:53

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

Yes sure, the array with all numbers is build and shuffled. On the other hand, hash solutions require more memory and more cpu power they build the same array but slower. A hash is a better choice, if you use only some number out of a large range. For really really much numbers a database is my best choice.

Boris

Re^4: How to not print the same random number

shotgunefx

created: 2006-03-04 06:08:56

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

That's what I figured. I haven't kept up lately, didn't know if some lazy evaluation had snuck in there ;)

-Lee
"To be civilized is to deny one's nature."

Re^3: How to not print the same random number

fishbot_v2

created: 2006-03-04 10:30:39

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

The question becomes "are you going to use all the numbers in that range?" If so, you have to remember them all regardless of your approach. If no, then you could save memory by just tracking what you use... but this then becomes more processor-expensive.

Re^4: How to not print the same random number

shotgunefx

created: 2006-03-04 12:36:54

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

On the plus side, once you've "bought it", the cost is pretty much constant.

-Lee
"To be civilized is to deny one's nature."

Re^3: How to not print the same random number

ysth

created: 2006-03-05 02:53:35

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

Not much of a hog. Storing the numbers already picked in a hash would be about as bad. (I'm not sure how many copies of the list the shuffle approach will end up having.)

But anyway, if you go with what I understand to be your idea of repeatedly generating random numbers until you find one not yet in a hash, you'll end up taking a very very long time finding the last few numbers if the limit is, say, 1000000.

Re^4: How to not print the same random number

shotgunefx

created: 2006-03-05 15:43:42

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

Well, I suppose I should have been more clear. I would think for many applications, the number of numbers picked compared to the number range would be relatively small.

So I guess the proper answer would be, "it depends".

-Lee
"To be civilized is to deny one's nature."

To make the concept clear, let me compare to a deck of cards. How do you prevent from drawing the same card twice? By not putting the card back in the deck before you draw another one.

You get the same result by simply shuffling the deck, and picking the next card, one by one.

shuffle is a function you can find in [cpan://List::Util]. You use it like this:

use List::Util 'shuffle';
my @shuffled_deck = shuffle 1 .. 10;
local $\ = "\n";
foreach(@shuffled_deck) {
    print;
}

1
2
9
8
7
10
4
5
6
3

It'll be far more efficient than retrying until you get something you haven't seen before, especially near the end, when you've almost seen them all.

use Data::Random qw(:all);
my @random_chars = rand_chars( set => 'numeric', min => 10, max => 10 );
foreach (@random_chars)
{print "$_\n";}

use Data::Random qw(:all);
my @random_chars = rand_chars( set => 'numeric', min => 10, max => 10 );
foreach (@random_chars)
{print "$_\n";}

This may give you what you want:

use strict;
use warnings;

print join"\n",grep$_,map{our%s;my$n=1+int rand(10);$s{$n}++?0:$n}1..100;

Re^2: How to not print the same random number

ysth

created: 2006-03-05 02:56:28

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

I like the "may" part. Did you calculate the actual odds?

Re^3: How to not print the same random number

GrandFather

created: 2006-03-05 04:41:28

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

Statistics is not my strong suit. I ran a test version a few times and noticed that about 23-25 itterations sufficed often enough and guessed that 100 would seldom dissapoint. Glad you noticed the "may". :)

---

DWIM is Perl's answer to Gödel

or you could just pull numbers out at random then throw them away:

use strict;
use warnings;

my @numbers = 1..10;
print splice (@numbers, rand (@numbers), 1), "\n" while @numbers;

If you're excluding numbers, it's not random. For example, if you've already chosen 9 numbers, the tenth is completely determined. FWIW.

@arr = 1..10;
randomize(\@arr);

print "@arr";

sub randomize {
    my ($p, $size, $swap, $key);
    $p = $_[0];

    $size = $#$p;
    for (0..($size-1)) {
        $swap = int rand($size - $_ + 1) + $_;
        $key = $p->[$_];
        $p->[$_] = $p->[$swap];
        $p->[$swap] = $key;
    }
}

Re^2: How to not print the same random number

Anonymous Monk

created: 2006-03-04 10:39:07

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

Hello.
Thanks for all the example.
If is not to much to ask but could you give me some explanation on what each codeline does?.

//Clyde

Re^3: How to not print the same random number

acid06

created: 2006-03-04 12:27:16

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

Using perlmonks for all your homework is how you expect to graduate?
Seriously, you better start learning because people won't always be this helpful.

/me readies himself for the possible -- to come

acid06
perl -e "print pack('h*', 16369646), scalar reverse $="

Re^2: How to not print the same random number

GrandFather

created: 2006-03-05 16:17:20

perlmonks

• reply
• node
• thread
• author

• node
• author
• thread

prlmnks

• node
• thread
• author

You can eliminate the manifest temporary by:

($p->[$_], $p->[$swap]) = ($p->[$swap], $p->[$_]);

Note though that it is slower:

use strict;
use warnings;
use Benchmark qw(cmpthese);

my @arr = 1..2;

cmpthese (-1,
            {
            temp => sub {my $temp = $arr[0]; $arr[0] = $arr[1]; $arr[1] = $temp;},
            swap => sub {($arr[0], $arr[1]) = ($arr[1], $arr[0]);},
            }
         );

          Rate swap temp
swap 1381331/s   -- -43%
temp 2425611/s  76%   --

---

DWIM is Perl's answer to Gödel

## Assign the numbers 1-10 to array @arr
@arr = 1..10;

## Pass @arr by reference to function randomize()
randomize(\@arr);

## When used in this context, the array prints
## with space delimiters between items
print "@arr";

sub randomize {
    ## Declare variables inside the scope of
    ## this function
    my ($p, $size, $swap, $key);

    ## Assign the array reference passed to
    ## the function to $p
    $p = $_[0];

    ## Assign the length of the array pointed
    ## to by $p to $size. This is the number
    ## of items in the array - 1
    $size = $#$p;

    ## Randomize all items up to the next to
    ## last item. The last item is randomized
    ## already if the other items are.
    for (0..($size-1)) {

        ## Pick a random item from current
        ## position to end
        $swap = int rand($size - $_ + 1) + $_;

        ## Swap item with current position
        $key = $p->[$_];
        $p->[$_] = $p->[$swap];
        $p->[$swap] = $key;
    }
}