removing a sublist

[Aldebaran]

When I study a syntax, I try to use it for any odd task that turns up. We have a contest for a club of sorts where the winner gets a prize of value $25 or so, but the idea is to foster participation. Hence, we don't want perpetual winners or the winner to be the person who sits at his computer all day, so we're choosing a random correct answer as opposed to the first.

Where I'm stuck now is making a composite of the those who answer correctly less those who have already won. I looked for examples in the books I have and couldn't find one. If I were to proceed in the manner of the syntaxes I've studied before, I would loop through indices, making exhaustive comparisons. What is the perl way to remove a sublist from another list

$ perl rand1.pl
7246281254938614119473194136957146644855362901505395617562308444592359
+441659779924940736387588266118823158415004985978328481067994672839164
[phone]
2 
$ cat rand1.pl
#!/usr/bin/perl -w
use strict;
use feature ':5.10';
use Crypt::Random qw( makerandom ); 

my $r = makerandom ( Size => 512, Strength => 1 ); 
print "$r\n";

srand($r);
my @winners=("ruth","sheila");

my @correct=("ruth","martin","dee","zack");

my @composite = @correct - @winners;

print "@composite \n";
$

Thanks in advance,

[Skeeve]

I always use a hash for this:

my @winners=("ruth","sheila");
my @correct=("ruth","martin","dee","zack");
my %correct;
@correct{@correct}= ();
delete @correct{@winners};
print join(' ',keys %correct),"\n";

[Aldebaran]

Alright, Skeeve, that's exactly the way I was headed, but it doesn't take me very long before I'm confounded by it. I find myself always wanting to revert to indices and crush it with brute force exhaustion like a numerical integration in fortran. I rarely find myself disappointed when I trust perl syntax.

Am I correct to think that the keys do not have to be unique? In my mind, the keys would represent the number of times a person has won. Stan won twice and dan won once. How do I marry @frequency to @correct in the script below?

$ perl tim1.pl
fan dango
4
4
2 1
$ cat tim1.pl
#!/usr/bin/perl -w
use strict;
use 5.10.0;
use Crypt::Random qw( makerandom );

# initialize pseudorandoms
my $r = makerandom( Size => 512, Strength => 1 );
# print "$r\n";
srand($r);

#populate lists
my @winners = qw/stan dan/;
my @winner_copy = @winners;
my @correct = qw/stan dan fan dango/;
my @correct_copy = @correct;

#populate hash
my %correct;
@correct{@correct} = ();
delete @correct{@winners};
print join( ' ', keys %correct ), "\n";

my $count = scalar(@correct);
print "$count\n";
my @new_list = @correct;
$count = scalar(@new_list);
print "$count\n";

# if ( scalar(@correct) eq 0 )
my @frequency = (2, 1);
# --@frequency;  that didn't work
print "@frequency", "\n";
# decrement frequency
# does anyone with a key value of zero have a correct response yet?
# if yes, you have a winner; if not then test whether there was a corr
+ect response among those with positive keys.
# if no one correct among them, then there was no correct response, be
+cause that partitions the sets.
$

I tried to map out in pseudoscript what I'm doing here. Obviously, I don't have the parts to make a control that works yet, but the contest just started today, so I don't have to have winner until after the time window closes.

Why does the hash have 4 nodes after 2 have been deleted?

How do I decrement the keys of a hash?

[eyepopslikeamosquito]

See perlfaq4 (arrays section) and Perl Cookbook, recipe 4.8 "Finding elements in one array but not another".

I would do it like this:

my @winners = ( "ruth", "sheila" );
my @correct = ( "ruth", "martin", "dee", "zack" );

# Build lookup table of winners
my %in_winners = map { $_ => 1 } @winners;

# Find items that are in @correct but not in @winners
my @correct_only = grep { not $in_winners{$_} } @correct;

[toolic]

Acme::Tools

use warnings;
use strict;
use Acme::Tools qw(minus);
use Data::Dumper;

my @winners=("ruth","sheila");
my @correct=("ruth","martin","dee","zack");
my @composite = minus(\@correct, \@winners);
print Dumper(\@composite);

__END__

$VAR1 = [
          'martin',
          'dee',
          'zack'
        ];

Also, qw is a convenient list constructor.

[Riales]

my @no_prev_winners = grep {
    my $correct = $_; !grep { $_ eq $correct } @winners
} @correct;

[Cristoforo]

List::Compare will also solve this problem.

#!/usr/bin/perl
use strict;
use warnings;
use 5.014;
use List::Compare;

my @winners=("ruth","sheila");
my @correct=("ruth","martin","dee","zack");

my $lc = List::Compare->new(\@winners, \@correct);

my @composite = $lc->get_Ronly;

print "@composite\n";

Prints
dee martin zack

Chris

[Kenosis]

Good solutions have been provided. The following includes randomly selecting a new winner:

map{$winners_hash{$_}++}@winners;
@new_winners = grep{!$winners_hash{$_}}@correct;
$the_winner = $new_winners[int(rand($#new_winners+1))];

Unless you're using an old version of perl, rand automatically calls srand.

Hope this helps!