in reply to Select three random numbers between 1..10

push @num, splice(@array, int rand @array, 1);

[download]
Which will remove and return a random element from the array, there by, never picking the same number twice.

Replies are listed 'Best First'.
Re: Re: Select three random numbers between 1..10
by Limbic~Region (Chancellor) on Mar 16, 2004 at 21:19 UTC
    Paladin,
    Wow. That certainly is much neater than my solution. I knew there were likely canned solutions, but it seemed like a cool problem to solve: 
    #!/usr/bin/perl
use strict;
use warnings;

print "$_\n" for Get_Unique_Random(10 , 3);

sub Get_Unique_Random {
    my ($max , $total) = @_;
    #  Requires error checking obviously
    $total ||= 3;
    my @return;
    while ( @return < $total ) {
         my @list = ( 1 .. $max );
         for my $seen ( @return ) {
             @list = grep { $_ ne $seen } @list;
         }
         push @return , $list[ (int rand $#list) + 1 ];
    }
    return wantarray ? @return : \@return;
}

    [download]
    Cheers - L~R
[reply]
[d/l]
Re: Re: Select three random numbers between 1..10
by hsinclai (Deacon) on Mar 18, 2004 at 02:46 UTC
    Honestly, 
    push @num, splice(@array, int rand @array, 1);

    [download]
    This totally made my day - the efficiency. In an extension of the original requirement, I needed to generate long fixed length strings from random elements in an array.. this is so much faster than what I was using.. I've no absolute requirement for total uniqueness, but it looks like no letter is being repeated in multiple runs of: 
    


#!/usr/bin/perl -w

use strict;

my $counter = "0";
my $randstring;
my @testarray = ( 'a' .. 'z', 'A' .. 'Z', '0' .. '9' );

do { $randstring .= splice(@testarray, int rand @testarray,1); 
     ++$counter;
} until $counter == 62; 

print "\n\$randstring:  $randstring\n\n";

    [download]
    -hsinclai
[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom