fdupe 查找重复文件的Perl脚本代码

代码如下:
#!/usr/bin/perl
#
# fdupe tool - finding duplicate files
#
# $Id: fdupe,v 1.7 2011/10/14 20:11:21 root Exp root $
#
# Source code Copyright (c) 1998,2011 Bernhard Schneider.
# May be used only for non-commercial purposes with
# appropriate acknowledgement of copyright.
#
# FILE :        fdupe
# DESCRIPTION : script finds duplicate Files.
# AUTHOR:       Bernhard Schneider <bernhard@neaptide.org>
# hints, crrections & ideas are welcome
#
# usage: fdupe.pl <path> <path> ...
#        find / -xdev | fdupe.pl
#
# how to select and remove duplicates:
#   redirect output to >file, edit the file and mark lines you
#   wish to move/delete with a preceding dash (-)
#   Use following script to delete marked files:
#   #!/usr/bin/perl -n
#   chomp; unlink if s/^-//;
#
# history:
# 12.05.99 - goto statment replaced with next
# 14.05.99 - minor changes
# 18.05.99 - removed confusing 'for $y'
#            included hash-search
# 20.05.99 - minor changes
# 02.03.00 - some functions rewritten, optimized for speed
# 10.01.01 - hint-fix by Ozzie |ozric at kyuzz.org|
# 05.03.02 - fixed hangups by reading block/char-Devices
# 08.09.11 - skips checking of hard links
# 14.10.11 - accept file names from stdin
#
#use strict; # uncomment for debugging

$|=1;
local (*F1,*F2); my %farray = (); my $statF1;

# ------------------------------
# traverse directories
sub scan ($) {
    my ($dir) = $_[0];
    opendir (DIR, $dir) or die "($dir) $!:$@";
    map {
          (-d) ? scan ($_) : push @{$farray{-s $_}},$_
             unless (-l or -S  or -p or -c or -b);
    } map "$dir/$_", grep !/^/./.?$/, readdir (DIR); closedir (DIR);
}

# ------------------------------
# get chunk of bytes from a file
sub getchunk ($$) {
  my ($fsize,$pfname) = @_;
  my $chunksize = 32;
  my ($nread,$buff);

  return undef unless open(F1,$$pfname);

  $statF1 = [(stat  F1)[3,1]];
  binmode F1;
  $nread = read (F1,$buff,$chunksize);
  ($nread == $chunksize || $nread == $fsize) ? "$buff" : undef;
} 

# ------------------------------
# compare two files
sub mycmp ($) {
  my ($fptr) = $_[0];
  my ($buffa, $buffb);
  my ($nread1,$nread2);
  my $statF2;
  my ($buffsize) = 16*1024;

  return -1 unless (open(F2,"<$$fptr"));

  $statF2 = [(stat  F2)[3,1]];

  return 0