List directory, print duplicated files names

print_duplicated.py

#!/usr/bin/env python

"""List directory, print duplicated files names
"""

from __future__ import print_function
import os
import argparse
import hashlib
import sys


def main(args):

    if not os.path.isdir(args.path):
        raise Exception('Not a directory')

    files = os.listdir(args.path)
    hashes = {}
    for file in files:
        filepath = os.path.join(args.path, file)
        if os.path.isfile(filepath):
            blocksize = 65536
            hasher = hashlib.md5()
            with open(filepath, 'rb') as afile:
                buf = afile.read(blocksize)
                while len(buf) > 0:
                    hasher.update(buf)
                    buf = afile.read(blocksize)
            filehash = hasher.hexdigest()
            if filehash in hashes.keys():
                hashes[filehash].append(file)
            else:
                hashes[filehash] = [file]
    out = ''
    for key, value in hashes.items():
        if len(value) > 1:
            for index, name in enumerate(value):
                separator = ' ' if index % 2 == 0 else '\n'
                out = out + name + separator
            out = out + '\n#####\n'
    print(out)


if __name__ == '__main__':
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument('-p', '--path', dest='path', type=str, help='Path look for duplicates')
    try:
        args = parser.parse_args()
        main(args)
    except:
        parser.print_help()
        sys.exit(0)