prepend bytes from https://bitcoin.stackexchange.com/users/25088/christopher-gurnee

pb.py

#!/usr/bin/env python3

from math import log, ceil
import sys


# lookup tables to convert integers in the range [0, 58) to base-58 digits and back
int_to_b58_digit = '123456789ABCDEFGHJKLMNPQRSTUVWXYZabcdefghijkmnopqrstuvwxyz'
b58_digit_to_int = { b58:i for i,b58 in enumerate(int_to_b58_digit) }

# convert a (long) integer to its base-58 representation string
def int_to_base58_str(int_rep):
    base58_str = ''
    while int_rep:
        int_rep, remainder = divmod(int_rep, 58)
        base58_str = int_to_b58_digit[remainder] + base58_str
    return base58_str


def prepended_bytes(prepended_b58_digits, b256_digit_count):

    # ones are a special case in base58check format;
    # count and remove them, they are added back in later
    ones = 0
    for b58 in prepended_b58_digits:
        if b58 != '1':
            break
        ones += 1
        prepended_b58_digits = prepended_b58_digits[1:]
    if not prepended_b58_digits:  # if they're all 1's
        return ones * b'\0'

    # calc the # of base58 digits required for b256_digit_count bytes of "real" data
    # (not including the prepended base58 digits)
    b58_digit_count = ceil(b256_digit_count * log(256) / log(58))

    do_overflow_check = True
    while True:
        # calc the minimum integral value which starts with the desired digits in base-58
        min_int = 0
        for b58 in prepended_b58_digits:
            min_int *= 58
            min_int += b58_digit_to_int[b58]
        #
        # left-shift (mult. by a power of 58) to be just left of the "real" base-58 data
        min_int *= 58 ** b58_digit_count

        # uncomment to sanity-check that min_int is correct
        #print(" min_int:", ones * '1' + int_to_base58_str(min_int))

        # right-shift by multiples of 8 bits (base-256) to retrieve only the
        # most-significant bytes which are to the left of the "real" base-256 data
        min_int >>= b256_digit_count * 8
        # right-shifing effectively rounds min_int down, but we
        # need it rounded up, so add one to round it up instead
        min_int += 1

        # because min_int has been rounded up above, it's possible that adding it to "real"
        # data could cause an overflow in base-58 making the prepended_b58_digits increment
        # by one; if this could happen, left-shift prepended_b58_digits and repeat
        if do_overflow_check:
            max_real_data_int = (1 << b256_digit_count*8) - 1
            max_base58_str = int_to_base58_str((min_int << b256_digit_count*8) + max_real_data_int)
            if not max_base58_str.startswith(prepended_b58_digits):
                prepended_b58_digits += '1'
                do_overflow_check = False  # it doesn't matter if the '1' appended above overflows to '2'

                # uncomment to confirm that the max possible value
                # wouldn't have the desired prepended base-58 digits
                #print("overflow:", ones * '1' + max_base58_str)

                continue

        # prepend any ones according to base58check rules, and convert min_int to a byte string
        return ones * b'\0' + min_int.to_bytes(length= (min_int.bit_length() + 7) // 8, byteorder='big')


if __name__ == '__main__':
    if len(sys.argv) != 3:
        sys.exit('usage: {} <STRING-TO-PREPEND> <DATA-BYTE-LEN (excluding 4-byte checksum)>'.format(sys.argv[0]))

    result = prepended_bytes(sys.argv[1], int(sys.argv[2]) + 4)  # add 4 for the checksum
    print(', '.join('{:#04x}'.format(i) for i in result))