Showing computing the tree hash on

chunked_hash.py

from hashlib import sha256
from typing import BinaryIO

CHUNK_LEN = 512 << 20  # 512 MB


def chunked_hash(data: BinaryIO, chunk_len: int = CHUNK_LEN) -> sha256:
    checksum = sha256()
    offset = 0

    while True:
        chunk = data.read(chunk_len)
        if not chunk:
            break

        chunk_checksum = sha256(chunk).hexdigest()
        checksum.update(f'{offset} {offset + len(chunk)} {chunk_checksum}\n'.encode())
        offset += len(chunk)

    return checksum