Be aware that Comma Separated Value "CSV" is far more complex in practice than simple test cases often suggest. This applies to Perl, Awk, whatever regex or parsing tool you are using.

If you are absolutely, positively, sure of your CSV format, you can write your own regex against it. Too often though, your CSV format is controlled by others and there are subtleties that will throw off your best laid plans.

Here is a regex I use often for CSV that I control. If I don't, I use Text::CSV cpan library...

That regex can be changed to get the nth field of CSV like so:

Code:

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

while(<DATA>) {
	chomp;
	my $str=$_;
	
	#if you want to deal with all at once:
	my @fields=/(?:^|,)("(?:[^"]+|"")*"|[^,]*)/g;
	my $i=0;
	print "All fields= $str\n";
	foreach my $field (@fields) {
   		print "field $i of $#fields = \'$field\'\n";
   		$i+=1;
   	}	
   	
   	#if you want to deal with field n
   	my $n=18;
   	my $actual=$n+1;  #there is no zeroth match...
   	$str=~/((?:^|,)(?:"(?:[^"]+|"")*"|[^,]*)){$actual}/; 	
   	print "single field, field $n=$1\n\n";
}

__DATA__
1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,"1-this is a test",20
1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,"2-this is a test",21

The beauty of Perl is the strength of the solutions on CPAN. In the case of CSV, XML, HTML, or other difficult to regex things, use a CPAN tool that has been tested in literally millions of cases so that you are more certain of your solution.

Here is a good overview of parsing CSV with Perl. This is the area of strength of Perl, and it still has its difficulties...