#!/usr/bin/perl

if($ARGV[0] =~ /\-a/i && !$ARGV[1]){
    system("find --version &>/dev/null");
    if($? != 0){
        print STDERR "FATAL: find not in \$PATH (".$ENV{'PATH'}.") - aborting.\n";
        exit(2);
    }
   
    $flist = `find ./`;
    @ARGV = split(/\n/, $flist);
}

if(@ARGV[0] =~ /\-(?:h|v|\?)/i || !$ARGV[0] || !$ARGV[1] || ($ARGV[0] =~ /\-a/i && $ARGV[1])){
    print STDERR "\nUsage: ".$0." <-a|files...>\n";
    print STDERR "\tAt least two or more files must be specified. Any folders\n";
    print STDERR "\tspecified will be skipped. Wildcards will work, for example:\n";
    print STDERR "\t\t".$0." *\n";
    print STDERR "\twill check all files in the current folder. Likewise:\n";
    print STDERR "\t\t".$0." folder/* folder-2/*\n";
    print STDERR "\twill check all files in both those folders. It will not\n";
    print STDERR "\trecurse on it's own.\n\n";
    print STDERR "\tAlternatively (as of 0.2a), you can pass the -a flag:\n";
    print STDERR "\t\t$0 -a\n";
    print STDERR "\tThis will check all files, recursively, starting from the\n";
    print STDERR "\tcurrent working directory.\n\n";
    print STDERR "\tAs of version 0.3a, it will also find files with duplicate\n";
    print STDERR "\tnames that reside in different directories. No flag is\n";
    print STDERR "\tnecessary to enable this functionality.\n\n";
    print STDERR $0." 0.3a by cmantito <http://code.cmantito.com>\n\n";
    exit(1);
}

system("md5sum --version &>/dev/null");
if($? != 0){
    print STDERR "FATAL: md5sum not in \$PATH (".$ENV{'PATH'}.") - aborting.\n";
    exit(2);
}

$progress = 0;
while($file = shift(@ARGV)){
    print "Processing ".$progress."/".++$#ARGV."...\n";
    if(-e $file){
        if( !-d $file){
            $escfile = $file;
            $escfile =~ s/(.)/\\$1/ig;
            $sum = `md5sum $escfile`;
            $sum =~ s/^(.+?)\s+.+$/\1/ig;
            chomp($sum);
            $sums{$sum}++;
            push(@$sum, $file);
           
            $filename = $file;
            $filename =~ s/(.+)\/(.+)$/\2/;
            if(!$filenames{$filename}){
                $filenames{$filename} = $file;
            }else{
                $filenames{$filename} .= "¬".$file;
            }
            $filenamecount{$filename}++;
        }
    }else{
        print STDERR "FATAL: Couldn't open file: ".$file." - aborting.\n";
        exit(1);
    }
    $progress++;
}

print "\n--- DUPLICATE FILES ---\n";
foreach $sum(keys %sums){
    if($sums{$sum} > 1){
        print $sum.": ";
        foreach $file(@$sum){
            print $file." ";
        }
        print "\n";
    }
}
print "\n--- DUPLICATE NAMES ---\n";
foreach $filename(keys %filenamecount){
    if($filenamecount{$filename} > 1){
        print $filename.": ";
        $filelist = $filenames{$filename};
        $filelist =~ s/¬/ /ig;
        print $filelist."\n";
    }
}

print "\n";
exit(0);