Get Git Object Hash

get_git_object_hash.py

import os
import hashlib

def get_git_object_hash(filepath):
    """
    If you use git hash-object <file>, it will give you git's 
    internal hash for that object: http://stackoverflow.com/a/552725/78202
    This method provides a way to get a git-compatible hash of the file
    at path `filepath`. 
    """

    # 64kb chunks
    BUF_SIZE = 65536

    filesize_bytes = os.path.getsize(filepath)

    sha1 = hashlib.sha1()
    sha1.update(("blob %u\0" % filesize_bytes).encode('utf-8'))

    with open(filepath, 'rb') as f:
        while True:
            data = f.read(BUF_SIZE)
            if not data:
                break
            sha1.update(data)

    sha1_str = sha1.hexdigest()

    # this should match git hash-object:
    # sha1_str_from_git = subprocess.check_output(['git', 'hash-object', filepath])
    # sha1_str_from_git = ''.join(sha1_str_from_git.split())
    # equal = sha1_str_from_git == sha1_str
    # logger.info("%s: subprocess check githash of file" % sha1_str_from_git)
    # logger.info("%s: computed githash of file" % sha1_str)
    # logger.info("Are they equal? %s" % equal)
    return sha1_str