Wayback Machine (web.archive.org) Digest generator

wayback-digest-gen.py

#!/usr/bin/env python3

import sys
import os
import hashlib
import base64

# Get the file name to generate the digest for

if len(sys.argv) < 2:
    print("Usage: %s <file>" % sys.argv[0])
    sys.exit(1)

filename = sys.argv[1]
if not os.path.isfile(filename):
    print("File %s not found" % filename)
    sys.exit(1)

# Calculate the SHA-1 of the file (https://gist.github.com/techtonik/df09baeacbebc52d234b)
sha1sum = hashlib.sha1()
with open(sys.argv[1], 'rb') as source:
  block = source.read(2**16)
  while len(block) != 0:
    sha1sum.update(block)
    block = source.read(2**16)


# Calculate the Base32 of the SHA-1
base32 = base64.b32encode(sha1sum.digest()).decode('utf-8').upper()

# Print the information
print("Filename: %s" % filename)
print("SHA-1: %s" % sha1sum.hexdigest())
print("Wayback Digest (Base32): %s" % base32)