megapy/mega/mega.py

587 lines
22 KiB
Python
Raw Normal View History

2013-02-04 02:02:33 +00:00
import re
import json
from Crypto.Cipher import AES
from Crypto.PublicKey import RSA
from Crypto.Util import Counter
import os
import random
import binascii
import requests
import time
import shutil
from .errors import ValidationError, RequestError
from .crypto import *
import tempfile
2013-02-04 02:02:33 +00:00
class Mega(object):
def __init__(self, options=None):
2013-02-04 02:02:33 +00:00
self.schema = 'https'
self.domain = 'mega.co.nz'
self.timeout = 160 # max time (secs) to wait for resp from api requests
2013-02-04 02:02:33 +00:00
self.sid = None
self.sequence_num = random.randint(0, 0xFFFFFFFF)
self.request_id = make_id(10)
2013-02-04 02:02:33 +00:00
if options is None:
options = {}
self.options = options
def login(self, email, password):
self.login_user(email, password)
return self
2013-02-04 02:02:33 +00:00
def login_user(self, email, password):
password_aes = prepare_key(str_to_a32(password))
uh = stringhash(email, password_aes)
resp = self.api_request({'a': 'us', 'user': email, 'uh': uh})
#if numeric error code response
if isinstance(resp, int):
raise RequestError(resp)
2013-02-04 02:02:33 +00:00
self._login_process(resp, password_aes)
def _login_process(self, resp, password):
encrypted_master_key = base64_to_a32(resp['k'])
self.master_key = decrypt_key(encrypted_master_key, password)
if 'tsid' in resp:
tsid = base64_url_decode(resp['tsid'])
key_encrypted = a32_to_str(
encrypt_key(str_to_a32(tsid[:16]), self.master_key))
if key_encrypted == tsid[-16:]:
self.sid = resp['tsid']
elif 'csid' in resp:
encrypted_rsa_private_key = base64_to_a32(resp['privk'])
rsa_private_key = decrypt_key(encrypted_rsa_private_key,
self.master_key)
2013-02-04 02:02:33 +00:00
private_key = a32_to_str(rsa_private_key)
self.rsa_private_key = [0, 0, 0, 0]
for i in range(4):
l = ((ord(private_key[0]) * 256 + ord(private_key[1]) + 7) / 8) + 2
2013-02-04 02:02:33 +00:00
self.rsa_private_key[i] = mpi_to_int(private_key[:l])
private_key = private_key[l:]
encrypted_sid = mpi_to_int(base64_url_decode(resp['csid']))
rsa_decrypter = RSA.construct(
(self.rsa_private_key[0] * self.rsa_private_key[1],
0L, self.rsa_private_key[2], self.rsa_private_key[0],
self.rsa_private_key[1]))
sid = '%x' % rsa_decrypter.key._decrypt(encrypted_sid)
sid = binascii.unhexlify('0' + sid if len(sid) % 2 else sid)
self.sid = base64_url_encode(sid[:43])
def api_request(self, data):
params = {'id': self.sequence_num}
self.sequence_num += 1
if self.sid:
params.update({'sid': self.sid})
#ensure input data is a list
if not isinstance(data, list):
data = [data]
2013-02-04 02:02:33 +00:00
req = requests.post(
'{0}://g.api.{1}/cs'.format(self.schema, self.domain),
params=params,
data=json.dumps(data),
timeout=self.timeout)
json_resp = json.loads(req.text)
2013-02-04 02:02:33 +00:00
#if numeric error code response
if isinstance(json_resp, int):
raise RequestError(json_resp)
2013-02-04 02:02:33 +00:00
return json_resp[0]
def parse_url(self, url):
#parse file id and key from url
if ('!' in url):
match = re.findall(r'/#!(.*)', url)
path = match[0]
return path
else:
raise RequestError('Url key missing')
def process_file(self, file, shared_keys):
"""
Process a file
"""
if file['t'] == 0 or file['t'] == 1:
keys = dict(keypart.split(':', 1) for keypart in file['k'].split('/') if ':' in keypart)
uid = file['u']
key = None
# my objects
if uid in keys:
key = decrypt_key(base64_to_a32(keys[uid]), self.master_key)
# shared folders
elif 'su' in file and 'sk' in file and ':' in file['k']:
shared_key = decrypt_key(base64_to_a32(file['sk']), self.master_key)
key = decrypt_key(base64_to_a32(keys[file['h']]), shared_key)
if file['su'] not in shared_keys:
shared_keys[file['su']] = {}
shared_keys[file['su']][file['h']] = shared_key
# shared files
elif file['u'] and file['u'] in shared_keys:
for hkey in shared_keys[file['u']]:
shared_key = shared_keys[file['u']][hkey]
if hkey in keys:
key = keys[hkey]
key = decrypt_key(base64_to_a32(key), shared_key)
break
if key is not None:
# file
if file['t'] == 0:
k = (key[0] ^ key[4], key[1] ^ key[5], key[2] ^ key[6],
key[3] ^ key[7])
file['iv'] = key[4:6] + (0, 0)
file['meta_mac'] = key[6:8]
# folder
else:
k = key
file['key'] = key
file['k'] = k
attributes = base64_url_decode(file['a'])
attributes = decrypt_attr(attributes, k)
file['a'] = attributes
# other => wrong object
elif file['k'] == '':
file['a'] = False
elif file['t'] == 2:
self.root_id = file['h']
file['a'] = {'n': 'Cloud Drive'}
elif file['t'] == 3:
self.inbox_id = file['h']
file['a'] = {'n': 'Inbox'}
elif file['t'] == 4:
self.trashbin_id = file['h']
file['a'] = {'n': 'Rubbish Bin'}
return file
def init_shared_keys(self, files, shared_keys):
"""
Init shared key not associated with a user.
Seems to happen when a folder is shared,
some files are exchanged and then the
folder is un-shared.
Keys are stored in files['s'] and files['ok']
"""
ok_dict = {}
for ok_item in files['ok']:
shared_key = decrypt_key(base64_to_a32(ok_item['k']), self.master_key)
ok_dict[ok_item['h']] = shared_key
for s_item in files['s']:
if s_item['u'] not in shared_keys:
shared_keys[s_item['u']] = {}
if s_item['h'] in ok_dict:
shared_keys[s_item['u']][s_item['h']] = ok_dict[s_item['h']]
2013-02-23 15:22:12 +00:00
##########################################################################
# GET
def find(self, filename):
"""
Return file object from given filename
"""
files = self.get_files()
for file in files.items():
if file[1]['a'] and file[1]['a']['n'] == filename:
return file
2013-02-04 02:02:33 +00:00
def get_files(self):
"""
Get all files in account
"""
2013-02-04 02:02:33 +00:00
files = self.api_request({'a': 'f', 'c': 1})
files_dict = {}
shared_keys = {}
self.init_shared_keys(files, shared_keys)
2013-02-04 02:02:33 +00:00
for file in files['f']:
processed_file = self.process_file(file, shared_keys)
#ensure each file has a name before returning
if processed_file['a']:
files_dict[file['h']] = processed_file
2013-02-04 02:02:33 +00:00
return files_dict
2013-02-07 23:38:05 +00:00
def get_upload_link(self, file):
"""
2013-02-07 23:47:10 +00:00
Get a files public link inc. decrypted key
Requires upload() response as input
"""
2013-02-07 23:30:37 +00:00
if 'f' in file:
file = file['f'][0]
public_handle = self.api_request({'a': 'l', 'n': file['h']})
file_key = file['k'][file['k'].index(':') + 1:]
decrypted_key = a32_to_base64(decrypt_key(base64_to_a32(file_key),
self.master_key))
return '{0}://{1}/#!{2}!{3}'.format(self.schema,
self.domain,
public_handle,
decrypted_key)
else:
raise ValueError('''Upload() response required as input,
use get_link() for regular file input''')
def get_link(self, file):
"""
Get a file public link from given file object
"""
file = file[1]
if 'h' in file and 'k' in file:
public_handle = self.api_request({'a': 'l', 'n': file['h']})
2013-04-19 18:38:45 +00:00
if public_handle == -11:
raise RequestError("Can't get a public link from that file (is this a shared file?)")
decrypted_key = a32_to_base64(file['key'])
return '{0}://{1}/#!{2}!{3}'.format(self.schema,
self.domain,
public_handle,
decrypted_key)
else:
raise ValidationError('File id and key must be present')
def get_user(self):
user_data = self.api_request({'a': 'ug'})
return user_data
def get_node_by_type(self, type):
"""
Get a node by it's numeric type id, e.g:
0: file
1: dir
2: special: root cloud drive
3: special: inbox
4: special trash bin
"""
nodes = self.get_files()
for node in nodes.items():
if (node[1]['t'] == type):
return node
2013-02-04 02:02:33 +00:00
def get_files_in_node(self, target):
"""
Get all files in a given target, e.g. 4=trash
"""
node_id = self.get_node_by_type(target)
files = self.api_request({'a': 'f', 'c': 1})
files_dict = {}
shared_keys = {}
self.init_shared_keys(files, shared_keys)
for file in files['f']:
processed_file = self.process_file(file, shared_keys)
if processed_file['a'] and processed_file['p'] == node_id[0]:
files_dict[file['h']] = processed_file
return files_dict
2013-02-05 07:11:15 +00:00
def get_id_from_public_handle(self, public_handle):
#get node data
#TODO fix this function
node_data = self.api_request({'a': 'f', 'f': 1, 'p': public_handle})
node_id = None
2013-02-04 02:02:33 +00:00
2013-03-10 03:53:24 +00:00
#determine node id
for i in node_data['f']:
if i['h'] is not u'':
node_id = i['h']
return node_id
2013-03-10 03:38:22 +00:00
def get_quota(self):
"""
2013-03-10 03:38:22 +00:00
Get current remaining disk quota in MegaBytes
"""
2013-03-10 03:38:22 +00:00
json_resp = self.api_request({'a': 'uq', 'xfer': 1})
#convert bytes to megabyes
return json_resp['mstrg'] / 1048576
2013-03-10 03:38:22 +00:00
def get_storage_space(self, giga=False, mega=False, kilo=False):
"""
Get the current storage space.
Return a dict containing at least:
'used' : the used space on the account
'total' : the maximum space allowed with current plan
All storage space are in bytes unless asked differently.
"""
if sum(1 if x else 0 for x in (kilo, mega, giga)) > 1:
raise ValueError("Only one unit prefix can be specified")
unit_coef = 1
if kilo:
unit_coef = 1024.
if mega:
unit_coef = 1048576.
if giga:
unit_coef = 1073741824.
json_resp = self.api_request({'a': 'uq', 'xfer': 1, 'strg': 1})
return {
'used': json_resp['cstrg'] / unit_coef,
'total': json_resp['mstrg'] / unit_coef,
}
2013-03-10 03:38:22 +00:00
def get_balance(self):
"""
2013-03-10 03:38:22 +00:00
Get account monetary balance, Pro accounts only
"""
user_data = self.api_request({"a": "uq", "pro": 1})
2013-03-10 03:38:22 +00:00
if 'balance' in user_data:
return user_data['balance']
##########################################################################
# DELETE
def delete(self, public_handle):
#straight delete by id
return self.move(public_handle, 4)
2013-02-04 04:42:28 +00:00
def delete_url(self, url):
#delete a file via it's url
path = self.parse_url(url).split('!')
2013-02-07 23:47:10 +00:00
public_handle = path[0]
file_id = self.get_id_from_public_handle(public_handle)
return self.move(file_id, 4)
def destroy(self, file_id):
#delete forever by private id
return self.api_request({'a': 'd',
'n': file_id,
'i': self.request_id})
def destroy_url(self, url):
#delete a file via it's url
path = self.parse_url(url).split('!')
public_handle = path[0]
file_id = self.get_id_from_public_handle(public_handle)
return self.destroy(file_id)
def move(self, file_id, target):
"""
Move a file to another parent node
params:
a : command
n : node we're moving
t : id of target parent node, moving to
i : request id
targets
2 : root
3 : inbox
4 : trash
"""
#determine target_node_id
target_node_id = str(self.get_node_by_type(target)[0])
return self.api_request({'a': 'm',
'n': file_id,
't': target_node_id,
'i': self.request_id})
def empty_trash(self):
# get list of files in rubbish out
files = self.get_files_in_node(4)
# make a list of json
if files != {}:
post_list = []
for file in files:
post_list.append({"a": "d",
"n": file,
"i": self.request_id})
return self.api_request(post_list)
##########################################################################
# DOWNLOAD
def download(self, file, dest_path=None):
"""
Download a file by it's file object
"""
self.download_file(None, None, file=file[1], dest_path=dest_path, is_public=False)
def download_url(self, url, dest_path=None):
"""
Download a file by it's public url
"""
path = self.parse_url(url).split('!')
file_id = path[0]
file_key = path[1]
self.download_file(file_id, file_key, dest_path, is_public=True)
2013-02-04 04:42:28 +00:00
def download_file(self, file_handle, file_key, dest_path=None, is_public=False, file=None):
if file is None :
if is_public:
file_key = base64_to_a32(file_key)
file_data = self.api_request({'a': 'g', 'g': 1, 'p': file_handle})
2013-04-19 18:38:45 +00:00
else:
file_data = self.api_request({'a': 'g', 'g': 1, 'n': file_handle})
k = (file_key[0] ^ file_key[4], file_key[1] ^ file_key[5],
file_key[2] ^ file_key[6], file_key[3] ^ file_key[7])
iv = file_key[4:6] + (0, 0)
meta_mac = file_key[6:8]
2013-04-19 18:38:45 +00:00
else:
file_data = self.api_request({'a': 'g', 'g': 1, 'n': file['h']})
k = file['k']
iv = file['iv']
meta_mac = file['meta_mac']
# Seems to happens sometime... When this occurs, files are
# inaccessible also in the official also in the official web app.
# Strangely, files can come back later.
2013-04-19 18:38:45 +00:00
if 'g' not in file_data:
raise RequestError('File not accessible anymore')
2013-02-04 02:02:33 +00:00
file_url = file_data['g']
file_size = file_data['s']
attribs = base64_url_decode(file_data['at'])
attribs = decrypt_attr(attribs, k)
file_name = attribs['n']
input_file = requests.get(file_url, stream=True).raw
if dest_path is None:
dest_path = ''
else:
dest_path += '/'
temp_output_file = tempfile.NamedTemporaryFile(mode='w+b', prefix='megapy_', delete=False)
2013-02-04 02:02:33 +00:00
counter = Counter.new(
128, initial_value=((iv[0] << 32) + iv[1]) << 64)
aes = AES.new(a32_to_str(k), AES.MODE_CTR, counter=counter)
file_mac = (0, 0, 0, 0)
for chunk_start, chunk_size in sorted(get_chunks(file_size).items()):
chunk = input_file.read(chunk_size)
chunk = aes.decrypt(chunk)
temp_output_file.write(chunk)
2013-02-04 02:02:33 +00:00
chunk_mac = [iv[0], iv[1], iv[0], iv[1]]
for i in range(0, len(chunk), 16):
block = chunk[i:i + 16]
2013-02-04 02:02:33 +00:00
if len(block) % 16:
block += '\0' * (16 - (len(block) % 16))
block = str_to_a32(block)
chunk_mac = [
chunk_mac[0] ^ block[0],
chunk_mac[1] ^ block[1],
chunk_mac[2] ^ block[2],
chunk_mac[3] ^ block[3]]
chunk_mac = aes_cbc_encrypt_a32(chunk_mac, k)
file_mac = [
file_mac[0] ^ chunk_mac[0],
file_mac[1] ^ chunk_mac[1],
file_mac[2] ^ chunk_mac[2],
file_mac[3] ^ chunk_mac[3]]
file_mac = aes_cbc_encrypt_a32(file_mac, k)
if self.options.get('verbose') is True:
# temp file size
file_info = os.stat(temp_output_file.name)
print('{0} of {1} downloaded'.format(file_info.st_size, file_size))
temp_output_file.close()
2013-02-04 02:02:33 +00:00
# check mac integrity
if (file_mac[0] ^ file_mac[1], file_mac[2] ^ file_mac[3]) != meta_mac:
raise ValueError('Mismatched mac')
shutil.move(temp_output_file.name, dest_path + file_name)
##########################################################################
# UPLOAD
2013-02-04 02:02:33 +00:00
def upload(self, filename, dest=None):
#determine storage node
if dest is None:
#if none set, upload to cloud drive node
2013-03-12 17:00:30 +00:00
if not hasattr(self, 'root_id'):
2013-02-04 02:02:33 +00:00
self.get_files()
dest = self.root_id
#request upload url, call 'u' method
input_file = open(filename, 'rb')
2013-04-28 15:33:50 +00:00
file_size = os.path.getsize(filename)
ul_url = self.api_request({'a': 'u', 's': file_size})['p']
2013-02-04 02:02:33 +00:00
#generate random aes key (128) for file
ul_key = [random.randint(0, 0xFFFFFFFF) for _ in range(6)]
count = Counter.new(128, initial_value=((ul_key[4] << 32) + ul_key[5]) << 64)
2013-02-04 02:02:33 +00:00
aes = AES.new(a32_to_str(ul_key[:4]), AES.MODE_CTR, counter=count)
file_mac = [0, 0, 0, 0]
2013-04-28 15:33:50 +00:00
upload_progress = 0
for chunk_start, chunk_size in sorted(get_chunks(file_size).items()):
2013-02-04 02:02:33 +00:00
chunk = input_file.read(chunk_size)
2013-04-28 15:33:50 +00:00
upload_progress += len(chunk)
2013-02-04 02:02:33 +00:00
#determine chunks mac
chunk_mac = [ul_key[4], ul_key[5], ul_key[4], ul_key[5]]
for i in range(0, len(chunk), 16):
block = chunk[i:i + 16]
2013-02-04 02:02:33 +00:00
if len(block) % 16:
block += '\0' * (16 - len(block) % 16)
block = str_to_a32(block)
chunk_mac = [chunk_mac[0] ^ block[0], chunk_mac[1] ^ block[1],
chunk_mac[2] ^ block[2],
2013-02-04 02:02:33 +00:00
chunk_mac[3] ^ block[3]]
chunk_mac = aes_cbc_encrypt_a32(chunk_mac, ul_key[:4])
#our files mac
file_mac = [file_mac[0] ^ chunk_mac[0], file_mac[1] ^ chunk_mac[1],
file_mac[2] ^ chunk_mac[2],
2013-02-04 02:02:33 +00:00
file_mac[3] ^ chunk_mac[3]]
file_mac = aes_cbc_encrypt_a32(file_mac, ul_key[:4])
#encrypt file and upload
chunk = aes.encrypt(chunk)
output_file = requests.post(ul_url + "/" + str(chunk_start),
data=chunk, timeout=self.timeout)
2013-02-04 02:02:33 +00:00
completion_file_handle = output_file.text
2013-04-28 15:33:50 +00:00
if self.options.get('verbose') is True:
# upload progress
print('{0} of {1} uploaded'.format(upload_progress, file_size))
2013-02-04 02:02:33 +00:00
#determine meta mac
meta_mac = (file_mac[0] ^ file_mac[1], file_mac[2] ^ file_mac[3])
attribs = {'n': os.path.basename(filename)}
encrypt_attribs = base64_url_encode(encrypt_attr(attribs, ul_key[:4]))
key = [ul_key[0] ^ ul_key[4], ul_key[1] ^ ul_key[5],
ul_key[2] ^ meta_mac[0], ul_key[3] ^ meta_mac[1],
2013-02-04 02:02:33 +00:00
ul_key[4], ul_key[5], meta_mac[0], meta_mac[1]]
encrypted_key = a32_to_base64(encrypt_key(key, self.master_key))
#update attributes
data = self.api_request({'a': 'p', 't': dest, 'n': [{
'h': completion_file_handle,
't': 0,
'a': encrypt_attribs,
'k': encrypted_key}]})
#close input file and return API msg
input_file.close()
return data
2013-03-12 14:40:56 +00:00
def create_folder(self, name, dest=None):
#determine storage node
if dest is None:
#if none set, upload to cloud drive node
if not hasattr(self, 'root_id'):
self.get_files()
dest = self.root_id
#generate random aes key (128) for folder
ul_key = [random.randint(0, 0xFFFFFFFF) for _ in range(6)]
2013-03-12 14:40:56 +00:00
#encrypt attribs
attribs = {'n': name}
encrypt_attribs = base64_url_encode(encrypt_attr(attribs, ul_key[:4]))
encrypted_key = a32_to_base64(encrypt_key(ul_key[:4], self.master_key))
#update attributes
data = self.api_request({'a': 'p',
't': dest,
'n': [{
'h': 'xxxxxxxx',
't': 1,
'a': encrypt_attribs,
'k': encrypted_key}
],
'i': self.request_id})
2013-03-12 14:40:56 +00:00
#return API msg
return data