hashlist.py

#!/usr/bin/env python
##
# directory tree containing file names and hashes
# github.com/deadbits
##
import sys
import os
import hashlib


def is_valid(file_path):
    if os.path.exists(file_path):
        if os.path.isfile(file_path):
            if os.path.getsize(file_path) > 0:
                return True
    return False


def get_hash(f, hash_type):
    fin = open(f, 'rb')
    if hash_type == 'md5':
        m = hashlib.md5()
    elif hash_type == 'sha1':
        m = hashlib.sha1()
    while True:
        data = fin.read(8192)
        if not data:
            break
        m.update(data)
    return m.hexdigest()


if __name__ == '__main__':
    try:
        root_dir = sys.argv[1]
        if not os.path.exists(root_dir):
            print 'error: path %s does not exist' % root_dir
            sys.exit(1)
        if os.path.isfile(root_dir):
            print 'error: %s is not a directory' % root_dir
            sys.exit(1)
    except IndexError:
        print 'usage: ./hashdeep.py <root directory>'
        sys.exit(0)

    # http://stackoverflow.com/questions/9727673/list-directory-tree-structure-using-python
    for root, dirs, files in os.walk(root_dir):
        level = root.replace(root_dir, '').count(os.sep)
        indent = ' ' * 4 * (level)
        print '%s%s/' % (indent, os.path.basename(root))
        subindent = ' ' * 4 * (level + 1)
        for fpath in [os.path.join(root, f) for f in files]:
            md5 = get_hash(fpath, 'md5')
            name = os.path.relpath(fpath, root_dir)
            print '%s%s\n%sMD5: %s\n' % (subindent, name, subindent, md5)