foreach my $e (@array1, @array2) {
    $union{$e}++ && $intersect{$e}++
}
[download]

my @sets = (\@array1, \@array2, ...);
for my $i (0..$#sets) {
  for my $j ($i+1..$#sets) {
    compute_stuff($sets[$i], $sets[$j]);
  }
}
[download]

sub union_intersection {
  my ($set1, $set2) = @_;
  my (%union, %intersection)
  for my $e (@$set1, @$set2) { ... }
  ...
}
[download]

I tired this first and am running into an error. I am very new at Perl so may be missing how the subroutine is set up. I run my script witht he following (my actual arrays have many more words than that in my sample):

my @sets = (\@array1, \@array2, \@array3, \@array4 );
for my $h (0..$#sets) {
    for my $j ($h+1..$#sets) {
        compute_stuff($sets[$h], $sets[$j]);
     }
}

Sub union_intersection {
    my ($set1, $set2) = @_;
    my (%union, %intersect);
        foreach my $e (@$set1, @$set2) {
            $union{$e}++ && $intersect{$e}++
        }
    my @intersect = sort keys %intersect;
    #print FILEOUT "@intersect\n";
    print FILEOUT scalar @intersect;
    print FILEOUT "\n";
}
[download]

I get a syntax error on 2 lines. One on this line: my (%union, %intersect); The other near the last line (with '}'). Not sure why. In you example you did't have the semi-colon at the end and added it but that wasn't it. Also, don't I need to call the subroutine somewhere for it to work. Not sure if this an error on my part (e.g. not sure what compute_stuff should be doing). Thanks for the help-

Your exact spec and your example program don't quite cooperate with one another, but this should get you closer:

Update: This code is buggy so I've put it in a readmore (not trying to hide that I make mistakes, after all). See Re^3: Array name with a Variable for the corrected code. The declaration of the hashes here is in the wrong scope and they weren't getting emptied properly.

If you don't want the intersection of an array and itself, there's an easy way to do that with flow control for the loops. You might also want to look into Set::Scalar or other CPAN modules that have done much of the work for you.

my ( %union, %intersect ) = ( (), () );

That seems a little extravagant. Do you often have problems with freshly-declared hashes not being empty?

No, I don't have that problem. It is superfluous. For some reason I was in the mindset that the example of how to assign to more than one variable might be useful to the OP, but I should've done that with @array1 through @array4 if I wanted to make that point I guess.

I do actually want to keep the intersection of the array to itself. I also tried this in my script and ran into an issue. I then tried it against the basic example to see if I can follow the steps. The output is as follows:

c d e
3
c d e
3
c d e
3
c d e
3
e
1
e f g h
4
e f g h
4
e f g h
4
d
1
....
[download]

I thought it would be more like this:

cde
3
e
1
d
1

0
e
1
....
[download]

I thought it would compare @array1 to @arrays1 - 4 then compare @array2 to @arrays1 - 4 and so forth. I think I am just not following the loops. Thanks-

Actually, I think I have a bug. Let me look a little closer. Yes, I have a bug. I moved the declaration of the variables into the wrong scope. This should work a little better.

use strict;
use warnings;

my @array1 = qw( c d e );
my @array2 = qw( e f g h );
my @array3 = qw( a b d );
my @array4 = qw( s g h j k l );

open my $out, '>', 'output' or die "Cannot open: $!\n";

foreach my $a1 ( \@array1, \@array2, \@array3, \@array4 ) {
    foreach my $a2 ( \@array1, \@array2, \@array3, \@array4 ) {
        my ( %union, %intersect );
        foreach my $item1 ( @$a1 ) {
            foreach my $item2 ( @$a2 ) {
                # print "item1: $item1\nitem2: $item2\n";
                $union{$item1}++;
                $union{$item2}++;
                if ( $item1 eq $item2 ) {
                    $intersect{$item1}++;
                }
            }
        }
        my @intersect = sort keys %intersect;
        print $out "@intersect\n"; #prints intersecting words
        print $out scalar @intersect . "\n";
    }
}

close $out or die "Error closing output; $!\n";
[download]

Is something like this what you are after?

use strict;
use warnings;
use Set::Scalar;

my @array1 = qw(c d e);
my @array2 = qw(e f g h);
my @array3 = qw(a b d);
my @array4 = qw(s g h j k l);
my @metaArray = (\@array1, \@array2, \@array3, \@array4, );

for my $first (0 .. $#metaArray - 1) {
    for my $second ($first + 1 .. $#metaArray) {
        my (%union,  %intersect);
        $union{$_}++ && $intersect{$_}++
            foreach @{$metaArray[$first]}, @{$metaArray[$second]};
        
        my @intersect = sort keys %intersect;
        print "\@array$first, \@array$second: @intersect - ";
        print scalar @intersect, "\n";
    }
}
[download]

Prints:

@array0, @array1: e - 1
@array0, @array2: d - 1
@array0, @array3:  - 0
@array1, @array2:  - 0
@array1, @array3: g h - 2
@array2, @array3:  - 0
[download]

If one of the input arrays has a duplicate entry, a spurious intersection will be detected.

An input array set of

my @array1 = qw(c d e);
my @array2 = qw(e f g h);
# my @array3 = qw(a b d);  # original code
my @array3 = qw(a a b d);  # modified for test
my @array4 = qw(s g h j k l);
[download]

Produces output of

@array0, @array1: e - 1
@array0, @array2: a d - 2
@array0, @array3:  - 0
@array1, @array2: a - 1
@array1, @array3: g h - 2
@array2, @array3: a - 1
[download]

godevars may be dealing with input arrays that have unique elements; certainly, his example data has this characteristic. However, it is nowhere stated explicitly that this is the case, so I just thought I'd mention it.