Perl takes a lot from `sed` already, so no need to shell out.

Without seeing your data it's hard to tell what you're trying to do exactly, but I'll assume you want to replace the comma at the end of the string.

Replace:

`sed 's/\(.*\),/\1/'`;
[download]

with...

s/,$//;
[download]

If that doesn't do what you want, supply a bit of the input file, and what your expected output should look like. Put both within <code></code> tags as you have with your code.

Also explain what you're trying to do with sort as well, as that can be avoided too.

What I want to do is remove the last comma in the file: for example:

21112,/vol/voly,blx
21113,/vol/eng,blz
21114,/vol/eng,file
[download]

I want the second comma to be gone. It is not at the end of the line.

If you have Perl version 5.10 or later, this also works:

c:\@Work\Perl\monks>perl -wMstrict -le
"use 5.010;
 ;;
 for my $s ('21112,/vol/voly,blx',  '21113,/vol/eng,blz',  '21114,/vol
+/eng,file') {
   my $t = $s;
   $t =~ s{ .* \K , }{}xms;
   print qq{'$s' -> '$t'};
   }
"
'21112,/vol/voly,blx' -> '21112,/vol/volyblx'
'21113,/vol/eng,blz' -> '21113,/vol/engblz'
'21114,/vol/eng,file' -> '21114,/vol/engfile'
[download]

Update: Slightly changed code example to enhance clarity.

---

Give a man a fish:  <%-{-{-{-<

G'day novice2015,

TMTOWTDI:

#!/usr/bin/env perl

use strict;
use warnings;

while (my $init_line = <DATA>) {
    my $neg_class_line = $init_line;
    my $neg_assert_line = $init_line;

    $neg_class_line  =~ s/,([^,]*)$/$1/;
    $neg_assert_line =~ s/,(?!.*,)//;

    print 'Initial line:       ', $init_line;
    print 'Negated class:      ', $neg_class_line;
    print 'Negative assertion: ', $neg_assert_line;
    print '-' x 40, "\n";
}

__DATA__
21112,/vol/voly,blx
21113,/vol/eng,blz
21114,/vol/eng,file
[download]

Output:

Initial line:       21112,/vol/voly,blx
Negated class:      21112,/vol/volyblx
Negative assertion: 21112,/vol/volyblx
----------------------------------------
Initial line:       21113,/vol/eng,blz
Negated class:      21113,/vol/engblz
Negative assertion: 21113,/vol/engblz
----------------------------------------
Initial line:       21114,/vol/eng,file
Negated class:      21114,/vol/engfile
Negative assertion: 21114,/vol/engfile
----------------------------------------
[download]

See perlre for a description of these, and other, methods.

Use Benchmark to find the most efficient method (if that's important to you).

And finally, a quick, friendly note on being specific. In one place you say "last comma" and in another you say "second comma": in the current (or future) data, these may not be the same. My two examples operate on the last comma (or change nothing if no commas are present). Adding these additional DATA lines:

W,X,Y,Z
W,X,YZ
W,XYZ
WXYZ
[download]

I get this additional output:

Initial line:       W,X,Y,Z
Negated class:      W,X,YZ
Negative assertion: W,X,YZ
----------------------------------------
Initial line:       W,X,YZ
Negated class:      W,XYZ
Negative assertion: W,XYZ
----------------------------------------
Initial line:       W,XYZ
Negated class:      WXYZ
Negative assertion: WXYZ
----------------------------------------
Initial line:       WXYZ
Negated class:      WXYZ
Negative assertion: WXYZ
----------------------------------------
[download]

— Ken

Gotcha... then this will look really similar. Just replace the whole sed line with: s/(.*),/$1/;