Torrent hardlink duplicator

torrent_hardlink.py

import io
import os
import sys
import json
import pathlib

def bencode_read(data):
    type_ = data.read(1)
    char = b""
    buf = b""
    if type_ == b'i':
        while 1:
            char = data.read(1)
            if char == b'e':
                break
            buf += char
        return int(buf)
    elif type_ in [b'l', b'd']:
        buf = []
        while 1:
            char = data.read(1)
            if char == b'e':
                break
            data.seek(data.tell()-1)
            buf.append(bencode_read(data))
        if type_ == b'l':
            return buf;
        else:
            return dict(zip(
                [x.decode() for x in buf[::2]],
                buf[1::2]
            ))
    else:
        data.seek(data.tell()-1)
        while 1:
            char = data.read(1)
            if char == b':':
                break
            buf += char
        return data.read(int(buf))

def parse_files(data):
    return {b'/'.join(x['path']).decode(): x['length']  for x in data}

check_dir = os.path.abspath(sys.argv[1])
torrent = sys.argv[2]

data = bencode_read(io.BufferedReader(open(torrent, 'rb')))

target_dir = os.path.abspath(data['info']['name'].decode())
os.mkdir(target_dir)

if data['info'].get('meta version') == 2:
    print("Torrent v2 detected, not supported")
    exit()

files = parse_files(data['info']['files'])

for path, target_size in files.items():
    fullpath = os.path.join(check_dir, path)
    print(fullpath)
    targetpath = os.path.join(target_dir, path)
    try:
        size = os.path.getsize(fullpath)
        if size == target_size:
            print("size match, linking")
            pathlib.Path(os.path.dirname(targetpath)).mkdir(parents=True, exist_ok=True)
            os.link(fullpath, targetpath)
    except Exception as e:
        print(e)
        print("file does not exist!")