#!/usr/bin/perl
#
# Cluester v0.1, written Hexacorn.com, 2012-05
#
# This is a simple script that attempts to divide list of files
# into clusters. Each cluster is based on a timestamp.
# Each timestamp is normalized to 1 minute (60 seconds),
# so that grouping is easier (and it has a wider range).
#
# Usage:
#      perl cluester.pl file_list_from_fls
#
# list from fls can be obtained via
#       fls -lrm <drive> -i <type> -f <type> \\.\<drive> > <output>
#    e.g.
#       fls -lrm f: -i raw -f ntfs \\.\f:  > list
#


use strict;
use warnings;

$| = 1;

print STDERR "
===================================================================
 Cluester v0.1, written by \@ Hexacorn.com, 2012-05
===================================================================
";

my $target = shift or die "\n\nError: Gimme a filename!\n";

if (!-f $target)
   {
      print "\n\nError: Don't know what to do with '$target'!\n";
      exit(-1);
   }


print STDERR "Processing \"$target\" file... \n";


my $file_size     = -s $target;
my $progress      = 0;
my $progress_temp = 0;
my %filesystem;

open    F, "<$target" || die "Can't open $target!\n";
binmode F;
while (<F>)
{
   s/[\r\n]+//g;
   $progress      += length($_);
   $progress_temp += length($_);
   if ($progress_temp > 65536)
   {
     print STDERR int(100*$progress/$file_size)."%\r";
     $progress_temp = 0;
   }

   # $row[0]  = MD5
   # $row[1]  = name
   # $row[2]  = inode
   # $row[3]  = mode_as_string
   # $row[4]  = UID
   # $row[5]  = GID
   # $row[6]  = size
   # $row[7]  = atime
   # $row[8]  = mtime
   # $row[9]  = ctime
   # $row[10] = crtime
   my @row    = split (/\|/);
   my $path   = $row [1];
   my $crtime = $row [10];
   my $GMT    = epoch2GMT ($crtime);
   my $GMT2   = epoch2GMT (60*int($crtime/60));

   $filesystem{$GMT2}[0]++;
   $filesystem{$GMT2}[ $filesystem{$GMT2}[0] ]="$GMT   $crtime   $path";

#   print STDERR "$ctime\t$path\n";
}
close F;

print STDERR "100%\n";

for my $k (sort keys %filesystem)
{
   my $prefix='';
   if ($filesystem{$k}[0]>=10)
   {
     $prefix = ' '  x 22;
   }

   for (my $i=0; $i<$filesystem{$k}[0]; $i++)
   {
     print "$prefix$k   $filesystem{$k}[1+$i]\n";
   }
   print "\n";

}

sub epoch2GMT
    {
      my $t = shift;

      my ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = localtime($t);

      $year += 1900;
      $mon  += 1;

      return sprintf ("%d-%02d-%02d %02d:%02d:%02d", $year,$mon,$mday,$hour,$min,$sec);

    }