Why not just parse each workbook of the spreadsheet using Spreadsheet::ParseExcel? The first example shows how to do this. Modifying this example to work on a per column basis (rather than by row) would be trivial.

Update: My mistake, you're working with a csv file.

book.csv:

1,2 ,3
1,2, 3
1,2,43
1
12,1
apple, 1, 3
orange
7,onion, 8, 9
[download]

Sample code using Text::CSV:

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

my $csv = Text::CSV->new({ sep_char => ',' });
 
my $file = "book.csv";

open(my $csvdata, '<', $file) 
    or die "Could not open '$file' $!\n";

while (my $line = <$csvdata>) {
    chomp $line;

    if ($csv->parse($line)) {     
        my @fields = $csv->fields();
        print Dumper \@fields;
    }
}
[download]

What exactly are you trying to achieve?