Last active
August 29, 2015 14:10
-
-
Save snnwolf/bb6a059e60a7a6545164 to your computer and use it in GitHub Desktop.
Скрипт для скачивания музыки из vk.com. 2 варианта: 1) через нити и очереди; 2) через нити и семафоры. Пожалуй с очередями правильнее!!!
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python | |
# -*- coding: utf-8 -*- | |
""" | |
Скрипт для скачивания музыки с сайта vkontakte.ru (vk.com) | |
Запуск: | |
python vkcom_audio_download.py | |
Принцип работы: | |
Скрипт проверяет сохраненный access_token. Если его нет или срок истек, | |
то открывается страница в браузере с запросом на доступ к аккаунту. | |
После подтверждения идет редирект на https://oauth.vk.com/blank.htm#... . | |
Нужно скопировать весь url, на который вас редиректнуло и вставить его | |
в консоль скрипта. | |
Далее будут скачиваться все ваши аудиозаписи. Если аудиозапись уже есть на | |
диске - то скачивания не происходит. | |
Будут запрошены ваши данные приложением с app_id = 3358129 | |
Можно создать свое Standalone-приложение с доступом к аудио здесь: | |
http://vk.com/editapp?act=create | |
И заменить APP_ID на ваше. | |
""" | |
import webbrowser | |
import pickle | |
import json | |
import urllib | |
import urllib2 | |
import HTMLParser | |
import re | |
import os | |
import urlparse | |
from pprint import pprint | |
import threading | |
import Queue | |
from datetime import datetime, timedelta | |
import time | |
import logging | |
logging.basicConfig(level=logging.DEBUG, format='[%(levelname)s] (%(threadName)-10s) %(message)s') | |
# id of vk.com application, that has access to audio | |
APP_ID = '4650301' | |
# if None, then save mp3 in current folder | |
MUSIC_FOLDER = None | |
# file, where auth data is saved | |
AUTH_FILE = '.auth_data' | |
# chars to exclude from filename | |
FORBIDDEN_CHARS = '/\\\?%*:|"<>!' | |
# кол-во потоков | |
THREAD_COUNT = 4 | |
def get_saved_auth_params(): | |
access_token = None | |
user_id = None | |
try: | |
with open(AUTH_FILE, 'rb') as pkl_file: | |
token = pickle.load(pkl_file) | |
expires = pickle.load(pkl_file) | |
uid = pickle.load(pkl_file) | |
if datetime.now() < expires: | |
access_token = token | |
user_id = uid | |
except IOError: | |
pass | |
return access_token, user_id | |
def save_auth_params(access_token, expires_in, user_id): | |
expires = datetime.now() + timedelta(seconds=int(expires_in)) | |
with open(AUTH_FILE, 'wb') as output: | |
pickle.dump(access_token, output) | |
pickle.dump(expires, output) | |
pickle.dump(user_id, output) | |
def get_auth_params(): | |
auth_url = ("https://oauth.vk.com/authorize?client_id={app_id}" | |
"&scope=audio&redirect_uri=http://oauth.vk.com/blank.html" | |
"&display=page&response_type=token".format(app_id=APP_ID)) | |
webbrowser.open_new_tab(auth_url) | |
redirected_url = raw_input("Paste here url you were redirected:\n") | |
aup = urlparse.parse_qs(redirected_url) | |
aup['access_token'] = aup.pop( | |
'https://oauth.vk.com/blank.html#access_token') | |
save_auth_params(aup['access_token'][0], aup['expires_in'][0], | |
aup['user_id'][0]) | |
return aup['access_token'][0], aup['user_id'][0] | |
def get_tracks_metadata(access_token, user_id): | |
url = ("https://api.vkontakte.ru/method/audio.get.json?" | |
"uid={uid}&access_token={atoken}".format( | |
uid=user_id, atoken=access_token)) | |
audio_get_page = urllib2.urlopen(url).read() | |
return json.loads(audio_get_page)['response'] | |
def get_track_full_name(t_data): | |
html_parser = HTMLParser.HTMLParser() | |
full_name = u"{0}_{1}".format( | |
html_parser.unescape(t_data['artist'][:100]).strip(), | |
html_parser.unescape(t_data['title'][:100]).strip(), | |
) | |
full_name = re.sub('[' + FORBIDDEN_CHARS + ']', "", full_name) | |
full_name = re.sub(' +', ' ', full_name) | |
return full_name + ".mp3" | |
# def Thread(f): | |
# """" | |
# Декорат для скачивания | |
# Для ограничения можно использовать семафоры | |
# """ | |
# def _inside(*a, **k): | |
# thr = threading.Thread(target = f, args = a, kwargs = k) | |
# thr.start() | |
# return _inside | |
# @Thread | |
def download_track(t_url, t_name): | |
# with semaphore: | |
t_path = os.path.join(MUSIC_FOLDER or "", t_name) | |
if not os.path.exists(t_path): | |
logging.info(u"Downloading {0}".format(t_name)) | |
urllib.urlretrieve(t_url, t_path) | |
logging.info(u'Finish {0}'.format(t_name)) | |
class Consumer(threading.Thread): | |
def __init__(self, task_queue): | |
threading.Thread.__init__(self) | |
self.task_queue = task_queue | |
def run(self): | |
while True: | |
next_task = self.task_queue.get() | |
if next_task is None: | |
# Poison pill means we should exit | |
logging.info('Exiting') | |
break | |
# logger.info('%s: %s' % (proc_name, next_task) | |
t_url, t_name = next_task | |
download_track(t_url, t_name) | |
return | |
def main(): | |
access_token, user_id = get_saved_auth_params() | |
if not access_token or not user_id: | |
access_token, user_id = get_auth_params() | |
tracks = get_tracks_metadata(access_token, user_id) | |
if MUSIC_FOLDER and not os.path.exists(MUSIC_FOLDER): | |
os.makedirs(MUSIC_FOLDER) | |
semaphore = threading.Semaphore(4) | |
tasks = Queue.Queue() | |
jobs = [Consumer(tasks) for i in xrange(THREAD_COUNT)] | |
for w in jobs: | |
w.daemon = True | |
w.start() | |
for t in tracks: | |
t_name = get_track_full_name(t) | |
# download_track(t['url'], t_name, semaphore) | |
tasks.put((t['url'], t_name)) | |
for i in xrange(THREAD_COUNT): | |
tasks.put(None) | |
# http://stackoverflow.com/questions/11815947/cannot-kill-python-script-with-ctrl-c | |
try: | |
while threading.active_count() > 1: | |
print 'threads:', threading.active_count(), [n.name for n in threading.enumerate()] | |
time.sleep(5) | |
except KeyboardInterrupt: | |
print '' | |
logging.info("Adieu") | |
if __name__ == '__main__': | |
main() |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python | |
# -*- coding: utf-8 -*- | |
""" | |
Скрипт для скачивания музыки с сайта vkontakte.ru (vk.com) | |
Запуск: | |
python vkcom_audio_download.py | |
Принцип работы: | |
Скрипт проверяет сохраненный access_token. Если его нет или срок истек, | |
то открывается страница в браузере с запросом на доступ к аккаунту. | |
После подтверждения идет редирект на https://oauth.vk.com/blank.htm#... . | |
Нужно скопировать весь url, на который вас редиректнуло и вставить его | |
в консоль скрипта. | |
Далее будут скачиваться все ваши аудиозаписи. Если аудиозапись уже есть на | |
диске - то скачивания не происходит. | |
Будут запрошены ваши данные приложением с app_id = 3358129 | |
Можно создать свое Standalone-приложение с доступом к аудио здесь: | |
http://vk.com/editapp?act=create | |
И заменить APP_ID на ваше. | |
""" | |
import webbrowser | |
import pickle | |
import json | |
import urllib | |
import urllib2 | |
import HTMLParser | |
import re | |
import os | |
import urlparse | |
from pprint import pprint | |
import threading | |
import Queue | |
from datetime import datetime, timedelta | |
import time | |
import logging | |
logging.basicConfig(level=logging.DEBUG, format='[%(levelname)s] (%(threadName)-10s) %(message)s') | |
# id of vk.com application, that has access to audio | |
APP_ID = '4650301' | |
# if None, then save mp3 in current folder | |
MUSIC_FOLDER = None | |
# file, where auth data is saved | |
AUTH_FILE = '.auth_data' | |
# chars to exclude from filename | |
FORBIDDEN_CHARS = '/\\\?%*:|"<>!' | |
# кол-во потоков | |
THREAD_COUNT = 4 | |
def get_saved_auth_params(): | |
access_token = None | |
user_id = None | |
try: | |
with open(AUTH_FILE, 'rb') as pkl_file: | |
token = pickle.load(pkl_file) | |
expires = pickle.load(pkl_file) | |
uid = pickle.load(pkl_file) | |
if datetime.now() < expires: | |
access_token = token | |
user_id = uid | |
except IOError: | |
pass | |
return access_token, user_id | |
def save_auth_params(access_token, expires_in, user_id): | |
expires = datetime.now() + timedelta(seconds=int(expires_in)) | |
with open(AUTH_FILE, 'wb') as output: | |
pickle.dump(access_token, output) | |
pickle.dump(expires, output) | |
pickle.dump(user_id, output) | |
def get_auth_params(): | |
auth_url = ("https://oauth.vk.com/authorize?client_id={app_id}" | |
"&scope=audio&redirect_uri=http://oauth.vk.com/blank.html" | |
"&display=page&response_type=token".format(app_id=APP_ID)) | |
webbrowser.open_new_tab(auth_url) | |
redirected_url = raw_input("Paste here url you were redirected:\n") | |
aup = urlparse.parse_qs(redirected_url) | |
aup['access_token'] = aup.pop( | |
'https://oauth.vk.com/blank.html#access_token') | |
save_auth_params(aup['access_token'][0], aup['expires_in'][0], | |
aup['user_id'][0]) | |
return aup['access_token'][0], aup['user_id'][0] | |
def get_tracks_metadata(access_token, user_id): | |
url = ("https://api.vkontakte.ru/method/audio.get.json?" | |
"uid={uid}&access_token={atoken}".format( | |
uid=user_id, atoken=access_token)) | |
audio_get_page = urllib2.urlopen(url).read() | |
return json.loads(audio_get_page)['response'] | |
def get_track_full_name(t_data): | |
html_parser = HTMLParser.HTMLParser() | |
full_name = u"{0}_{1}".format( | |
html_parser.unescape(t_data['artist'][:100]).strip(), | |
html_parser.unescape(t_data['title'][:100]).strip(), | |
) | |
full_name = re.sub('[' + FORBIDDEN_CHARS + ']', "", full_name) | |
full_name = re.sub(' +', ' ', full_name) | |
return full_name + ".mp3" | |
def Thread(f): | |
"""" | |
Декоратор для скачивания. | |
Для ограничения можно использовать семафоры. | |
Все процессы демонизируются. Поэтому в основной программе нужно добавить | |
while threading.active_count() > 1: | |
time.sleep(1) | |
""" | |
def _inside(*a, **k): | |
thr = threading.Thread(target = f, args = a, kwargs = k) | |
thr.daemon = True | |
thr.start() | |
return _inside | |
@Thread | |
def download_track(t_url, t_name, semaphore): | |
with semaphore: | |
t_path = os.path.join(MUSIC_FOLDER or "", t_name) | |
if not os.path.exists(t_path): | |
logging.info(u"Downloading {0}".format(t_name)) | |
urllib.urlretrieve(t_url, t_path) | |
logging.info(u'Finish {0}'.format(t_name)) | |
def main(): | |
access_token, user_id = get_saved_auth_params() | |
if not access_token or not user_id: | |
access_token, user_id = get_auth_params() | |
tracks = get_tracks_metadata(access_token, user_id) | |
if MUSIC_FOLDER and not os.path.exists(MUSIC_FOLDER): | |
os.makedirs(MUSIC_FOLDER) | |
semaphore = threading.Semaphore(4) | |
for t in tracks: | |
t_name = get_track_full_name(t) | |
download_track(t['url'], t_name, semaphore) | |
# http://stackoverflow.com/questions/11815947/cannot-kill-python-script-with-ctrl-c | |
try: | |
while threading.active_count() > 1: | |
print 'threads:', threading.active_count() #, [n.name for n in threading.enumerate()] | |
time.sleep(1) | |
except KeyboardInterrupt: | |
print '' | |
print("Adieu") | |
else: | |
print('All files downloaded') | |
if __name__ == '__main__': | |
main() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment