Thin-down - Remove Files

• Tweet
• 

        Removes duplicate files from subdirectories.

#!/usr/bin/perl -w
#
#

use Digest::MD5 qw(md5_hex);
use POSIX qw(:termios_h);

my $fd_stdin = fileno(STDIN);
my $term     = POSIX::Termios->new();
$term->getattr($fd_stdin);
my $oterm     = $term->getlflag();

my $echo     = ECHO | ECHOK | ICANON;
my $noecho   = $oterm & ~$echo;

sub cbreak {
    $term->setlflag($noecho);  # ok, so i don't want echo either
    $term->setcc(VTIME, 1);
    $term->setattr($fd_stdin, TCSANOW);
}

sub cooked {
    $term->setlflag($oterm);
    $term->setcc(VTIME, 0);
    $term->setattr($fd_stdin, TCSANOW);
}

sub readkey {
    my $key = '';
    cbreak();
    sysread(STDIN, $key, 1);
    cooked();
    return $key;
}

END { cooked() }

my $dir = ".";
my %FILES;
my %INODE;
my %COUNT;
my %SUMS;

sub sum_file($) {

    my $file = shift;

    open(FILE, $file) or die "Can't open '$file': $!";
    binmode(FILE);

    my $md5 = Digest::MD5->new;
    while (<FILE>) {
        $md5->add($_);
    }
    close(FILE);
    return $md5->b64digest;
}

open (FIND, "find . -type f -exec stat -c '%i;%s;%n' {} \\;|") ||
    die "Can't run find: $!\n";
while(<FIND>) {
    chomp();
    (my $inode, $size, $filename) = split(/;/,$_);
    $FILES{$size}{$COUNT{$size}++} = $filename;
    $INODE{$filename} = $inode;
}
close(FIND);

opendir(DIR, $dir) || die "Can't open directory '$dir' $!\n";
@files = grep { /\./i && -f "$dir/$_" } readdir(DIR);
closedir DIR;

for my $file (sort @files) {

    my $size = (stat($file))[7];

    $file = "./$file";

    if ($COUNT{$size} >= 2) {

        #
        # Don't sum the same inode twice.
        #
        $SUMS{$INODE{$file}} = sum_file($file) unless($SUMS{$INODE{$file}});

        my $sum = $SUMS{$INODE{$file}};

        for my $match (1..$COUNT{$size}) {

            my $match_file = $FILES{$size}{$match-1};

            #
            # Don't compart with itself.
            #
            next if ($INODE{$file} == $INODE{$match_file});

            $SUMS{$INODE{$match_file}} = sum_file($match_file)
                unless($SUMS{$INODE{$match_file}});

                print "-$sum- $file\n";
            if ( $SUMS{$INODE{$match_file}} eq "$sum" ) {

                print "$INODE{$file} $SUMS{$INODE{$file}} $file\n",
                    "$INODE{$match_file} $SUMS{$INODE{$match_file}}",
                    " $match_file\n";

                $| = 1;
                print "Remove '$file'? [y|N] ";

                until (defined ($key = readkey())){};
                chomp($key);

                if ($key =~ /^y$/i) {

                    unlink($file);
                    print "removed `$file`\n";

                } elsif ($key =~ /^\003$/i) {

                    die "\n";

                } else {

                    print "\n";

                }

            }

        }
    }
}

Tags:
• Perl
• thin_tools