Compare two arrays

[Anonymous Monk]

Hi Monks!
I have two arrays, I need to find if the elements of the first array matchs any element on the second array. I am getting stuck once I have the account numbers into a hash:

#!/usr/bin/perl

use strict;
use warnings;
use Data::Dumper;

my @account = qw( 96329XY 7776fc 334457yg 2233oop 000001 8877666b);
my @file =  qw(8877666b.txt 2233oop.txt 334457yg.txt 7776fc.txt 96329X
+Y_K.txt);

my %seen;

for my $acc (@account) {

   next unless $acc =~ /^\d+/;
   my $key = $acc;
   
   push @{$seen{$key}}, $acc;
}

#print Dumper %seen;

for my $filename (@file) {

   next unless $filename =~ /^\d+/;
   my $key = $filename;
   ...
}

Thanks for looking!

[Eily]

You should uncomment print Dumper %seen; and write \%seen instead of %seen: you are probably not getting what you expect in your %seen hash.

You may need to use captures in your regex (it depends on weither you want to use the whole name or just the number) in which case you'll find everything you need in Extracting matches (hint: it may be next unless $acc =~ /(^\d+)/; my $key = $1;/). Once your keys are what you expect them to be, in the filename loop you just have to check exists $seen{$key} to know if there is a match. See exists

[Anonymous Monk]

Why can't this work?


...
for my $filename (@file) {

   next unless $filename =~ /^\d+/;

   if(grep { %seen{$_} =~ $filename} keys %seen) {
     print " $filename\n";
    } else {
      print "$filename is not in array\n";
    }
}

[Eily]

%seen{$_} is incorrect, you're only accessing one element (ie: a scalar) so you have to use the sigil $: $seen{$_}

Now, look at your dump of \%test, you'll see that the value in $seen{$_} is actually an array, not a string, unless you have changed your data structure since the first post.

Now your grep test (if I ignore your previous errors), applied to 96329XY and 96329XY_K.txt would turn into : "96329XY" =~ /96329XY_K.txt/, you probably have the wrong order. And you probably want to use quotemeta: grep { $filename =~ /\Q$_/ } keys %seen.

[Bloodnok]

TIMTOWTDI ... why not try ...

#!/usr/bin/perl

use strict;
use warnings;
use Data::Dumper;

my @account = qw( 96329XY 7776fc 334457yg 2233oop 000001 8877666b);
my @file =  qw(8877666b.txt 2233oop.txt 334457yg.txt 7776fc.txt 96329X
+Y_K.txt);

my $ac_re = join '|', grep /^\d+/, @account;

print join "\n", grep /$ac_re/, @file;

or similar ?

Just a (nother) thought :-)

[hippo]

It's an FAQ: How do I compute the intersection of two arrays?

[mr_mischief]

Untested.

#!/usr/bin/perl

use strict;
use warnings;
use Data::Dumper;

my @account = qw( 96329XY 7776fc 334457yg 2233oop 000001 8877666b);
my @file =  qw(8877666b.txt 2233oop.txt 334457yg.txt 7776fc.txt 96329X
+Y_K.txt);

my %seen;

for my $acc (@account) {
    next unless $acc =~ /^\d+/;
    $seen{ $acc }++; # I'm not sure why you were pushing onto an array
+ to test for existence.
}

#print Dumper %seen;

for my $filename (@file) {
    next unless $filename =~ /^(\d+.*?)\.txt$/;
    my $basename = $1;
    print "seen in account array: $basename\n" if exists $seen{ $basen
+ame }; 
}

[karlgoethebier]

I tried it using Set::Scalar:

#!/usr/bin/env perl                                                   
+                                            

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

my @account = qw( 96329XY 7776fc 334457yg 2233oop 000001 8877666b);

my @file = qw(8877666b.txt 2233oop.txt 334457yg.txt 7776fc.txt 96329XY
+_K.txt);

s/\.txt// for @file;

my $account_set = Set::Scalar->new(@account);

my $file_set = Set::Scalar->new(@file);

my $intersection = $account_set * $file_set;

$intersection->as_string_callback( sub { join qq(\n), sort $_[0]->elem
+ents } );

print qq($intersection\n);

__END__

karls-mac-mini:monks karl$ ./1102555.pl                               
+                                            
2233oop                                                               
+                                            
334457yg                                                              
+                                            
7776fc                                                                
+                                            
8877666b

Edit: Fixed wrong link to used module.

Regards, Karl

[thanos1983]

Hello Anonymous Monk,

I can see that everyone has all ready proposed answers and probably all of them are working. I just could not resist to the idea of trying also to create my own answer since I think it a useful topic comparing arrays.

For my answer I use Data::Compare module which provides you which can provide you with several different outputs when you compare arrays. I have also combine my answer with a perlre (regular expression) taken from Regex To Remove File Extension which makes your solution more generic and not only to *.txt extensions.

I would recommend to modify the title to something more appropriate (e.g. Comparing arrays with regex). This is because it produces confusion to people in future that are looking of simply comparing arrays without file extensions etc. etc.

Having said that, sample of working code with output:

#!/usr/bin/perl
use strict;
use warnings;
use Data::Dumper;

sub array_utils {

    use Array::Utils qw(:all);
    
    my @compare = ();
    my @account = qw( 96329XY 7776fc 334457yg 2233oop 000001 8877666b)
+;
    my @file =  qw(8877666b.txt 2233oop.txt 334457yg.txt 7776fc.txt 96
+329XY_K.txt);

    foreach my $sample (@file) {
    $sample =~ s/(.+)\.[^.]+$/$1/;
    push @compare,$sample;
    }

    # symmetric difference
    return my @diff = array_diff(@account, @compare);

    # intersection
    #return my @isect = intersect(@account, @compare);
    
    # unique union
    #return my @unique = unique(@account, @compare);
    
    # check if arrays contain same members
    if ( !array_diff(@account, @compare) ) {
    return "Arrays contain same members!";
    }
    
    # get items from array @account that are not in array @compare
    #return my @minus = array_minus( @account , @compare );

}

my @final = array_utils();

print Dumper(\@final);

__END__

$VAR1 = [
          '96329XY',
          '000001',
          '96329XY_K'
        ];

Play around by removing the # symbols and observe the output, maybe one of them fits more to your needs.