Recursively calculate a "directory md5sum" based on it's content filename, size and mtime

recursive_dir_md5.py

import os
import hashlib

# calc directory checksum
def calc_dir_sum(path):
        for root, dirs, files in os.walk(path):
                dir_hash = hashlib.md5()

                # calc checksum for each file
                # based on file name, size and mtime
                for f in files:
                        fullfn = os.path.join(root, f)
                        # skip symlinks
                        if os.path.islink(fullfn): continue
                        s = os.stat(fullfn)
                        fstr = "{0}/{1}/{2}".format(f, s.st_size, s.st_mtime)
                        dir_hash.update(fstr)
                        #print "{0}  {1}".format(fullfn, hashlib.md5(fstr).hexdigest())

                # calc checksum for each dir
                # by recursively calling calc_dir_sum()
                for d in dirs:
                        fulldn = os.path.join(root, d)
                        # skip symlinks
                        if os.path.islink(fulldn): continue
                        dir_hash.update(calc_dir_sum(fulldn))

                print "{0}  {1}".format(dir_hash.hexdigest(), root + '/')
                return dir_hash.hexdigest()

calc_dir_sum("/your/data/path/")