An implementation of SHA-1 in pure Python 3

sha1_demo.py

#!/usr/bin/env python3

''' sha1_demo

    An implementation of SHA-1 in pure Python 3

    Built from the pseudocode at
    https://en.wikipedia.org/wiki/SHA-1#SHA-1_pseudocode

    Written by PM 2Ring 2017.10.01
'''

# SHA-1 initialization constants
h_init = 0x67452301, 0xEFCDAB89, 0x98BADCFE, 0x10325476, 0xC3D2E1F0

# A bitmask of 32 '1' bits, so we can do
# 32 bit unsigned arithmetic in Python
mask = 0xffffffff

def leftrotate(x, n):
    ''' Left rotate 32 bit integer x by n bits '''
    return ((x << n) | (x >> (32 - n))) & mask

def sha1(s):
    # Add padding bits and length to message
    ml = len(s) << 3
    d = (447 - ml) % 512
    n = 1 << d
    nlen = (1 + d) >> 3
    s += n.to_bytes(nlen, 'big') + ml.to_bytes(8, 'big')

    # Initialise the hash
    h = list(h_init)

    # Process the message in successive 512-bit chunks
    for chunk in zip(*[iter(s)] * 64):
        # Break chunk into sixteen 32-bit big-endian words
        w = [int.from_bytes(u, 'big') for u in zip(*[iter(chunk)] * 4)]

        # Extend the sixteen 32-bit words into eighty 32-bit words
        for i in range(16, 80):
            x = (w[i-3] ^ w[i-8] ^ w[i-14] ^ w[i-16])
            w.append(leftrotate(x, 1))

        # Initialize hash value for this chunk
        a, b, c, d, e = h

        # Main loop
        for i in range(80):
            if 0 <= i <= 19:
                f = (b & c) | ((b ^ mask) & d)
                k = 0x5A827999
            elif 20 <= i <= 39:
                f = b ^ c ^ d
                k = 0x6ED9EBA1
            elif 40 <= i <= 59:
                f = (b & c) | (b & d) | (c & d)
                k = 0x8F1BBCDC
            elif 60 <= i <= 79:
                f = b ^ c ^ d
                k = 0xCA62C1D6

            a, b, c, d, e = (
                (leftrotate(a, 5) + f + e + k + w[i]) & mask,
                a, leftrotate(b, 30), c, d
            )

        # Add this chunk's hash to result so far
        h = [(u + v) & mask for u, v in zip(h, (a, b, c, d, e))]

    # Produce the final hash value
    return b''.join([u.to_bytes(4, 'big') for u in h])

def test(numtests=1000, random_seed=None):
    ''' Validate sha1 against the standard version in hashlib '''
    seed(random_seed)
    s = b''
    for i in range(numtests):
        h0 = std_sha1(s).digest()
        h1 = sha1(s)
        assert h0 == h1, i
        # Prepend a random byte to s
        s = bytes([randrange(256)]) + s
        if i % 100 == 0:
            print('  {}'.format(i), end='\r', flush=True)
    print('\nok')

def main():
    ''' Calculate SHA-1 hashes of all files named on the command
        line, or read from stdin if no file names are given
    '''
    if len(sys.argv) == 1:
        fname = '-'
        data = sys.stdin.buffer.read()
        h = sha1(data)
        print('{}  -'.format(h.hex()))
    else:
        for fname in sys.argv[1:]:
            try:
                data = Path(fname).read_bytes()
            except IOError as e:
                print(e)
            else:
                h = sha1(data)
                print('{}  {}'.format(h.hex(), fname))


if __name__ == '__main__':
    import sys
    from pathlib import Path
    from hashlib import sha1 as std_sha1
    from random import seed, randrange

    main()
    #test()