Maybe useful:

CGI's Vars returns multi-valued parameters as a packed string, separated by the "\0" (null) character.

If you know of a reason why one should not use Vars(), it would be great, if you could send me some info about that reason...

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

my $cgi = CGI->new();

my %p = $cgi->Vars;
# if you don't like the \0 delimiter, replace it by your own one; e.g.
+:
s/\0/[whatever I want]/g for values %p;

print Dumper \%p;

__END__

$ perl T.pl a=1 b=1 b=2 b=3 
$VAR1 = {
          'a' => '1',
          'b' => '1[whatever I want]2[whatever I want]3'
        };
[download]

Example how to count the number of "\n" in a string; see #How can I count the number of occurrences of a substring within a string?

#!/usr/bin/perl -l
use strict;
use warnings;

my $text = qq(I
have
multiple
lines
);

print $text =~ s/\n/\n/g;       # substitute
print $text =~ tr/\n//;         # transliterate

__END__
[download]

Example how to use Text::CSV with a CSV file containing fields with multi-lined fields:

#!/usr/bin/perl
use strict;
use warnings;
use Text::CSV;


my $csv_file = 'test.csv';

# binary => 1 : makes Text::CSV work on multi-lined fields
my $csv      = Text::CSV->new( { binary => 1 } );

#open my $fh, '<', $csv_file or die "$csv_file: $!\n";
# *DATA; instead of opening a separate file; using the data section at
+ the end of file
my $fh       = *DATA;

# read rows and store them in an array reference
while ( my $ref = $csv->getline( $fh ) ) {
    # dereference the reference and simply print the elements
    print "@$ref\n";
}

#close $fh or die "$csv_file: close failed: $!\n";


__DATA__
"column1","column2","column3"
"single1","single2","single3"
"single4","multiA
multiB",single5
[download]