logo

youtube-dl

[mirror] Download/Watch videos from video hostersgit clone https://hacktivis.me/git/mirror/youtube-dl.git

viki.py (14590B)

    

  1. # coding: utf-8
  2. from __future__ import unicode_literals
  4. import hashlib
  5. import hmac
  6. import json
  7. import time
  9. from .common import InfoExtractor
  10. from ..utils import (
  11.     ExtractorError,
  12.     int_or_none,
  13.     parse_age_limit,
  14.     parse_iso8601,
  15.     try_get,
  16. )
  19. class VikiBaseIE(InfoExtractor):
  20.     _VALID_URL_BASE = r'https?://(?:www\.)?viki\.(?:com|net|mx|jp|fr)/'
  21.     _API_URL_TEMPLATE = 'https://api.viki.io%s'
  23.     _DEVICE_ID = '112395910d'
  24.     _APP = '100005a'
  25.     _APP_VERSION = '6.11.3'
  26.     _APP_SECRET = 'd96704b180208dbb2efa30fe44c48bd8690441af9f567ba8fd710a72badc85198f7472'
  28.     _GEO_BYPASS = False
  29.     _NETRC_MACHINE = 'viki'
  31.     _token = None
  33.     _ERRORS = {
  34.         'geo': 'Sorry, this content is not available in your region.',
  35.         'upcoming': 'Sorry, this content is not yet available.',
  36.         'paywall': 'Sorry, this content is only available to Viki Pass Plus subscribers',
  37.     }
  39.     def _stream_headers(self, timestamp, sig):
  40.         return {
  41.             'X-Viki-manufacturer': 'vivo',
  42.             'X-Viki-device-model': 'vivo 1606',
  43.             'X-Viki-device-os-ver': '6.0.1',
  44.             'X-Viki-connection-type': 'WIFI',
  45.             'X-Viki-carrier': '',
  46.             'X-Viki-as-id': '100005a-1625321982-3932',
  47.             'timestamp': str(timestamp),
  48.             'signature': str(sig),
  49.             'x-viki-app-ver': self._APP_VERSION
  50.         }
  52.     def _api_query(self, path, version=4, **kwargs):
  53.         path += '?' if '?' not in path else '&'
  54.         app = self._APP
  55.         query = '/v{version}/{path}app={app}'.format(**locals())
  56.         if self._token:
  57.             query += '&token=%s' % self._token
  58.         return query + ''.join('&{name}={val}.format(**locals())' for name, val in kwargs.items())
  60.     def _sign_query(self, path):
  61.         timestamp = int(time.time())
  62.         query = self._api_query(path, version=5)
  63.         sig = hmac.new(
  64.             self._APP_SECRET.encode('ascii'),
  65.             '{query}&t={timestamp}'.format(**locals()).encode('ascii'),
  66.             hashlib.sha1).hexdigest()
  67.         return timestamp, sig, self._API_URL_TEMPLATE % query
  69.     def _call_api(
  70.             self, path, video_id, note='Downloading JSON metadata', data=None, query=None, fatal=True):
  71.         if query is None:
  72.             timestamp, sig, url = self._sign_query(path)
  73.         else:
  74.             url = self._API_URL_TEMPLATE % self._api_query(path, version=4)
  75.         resp = self._download_json(
  76.             url, video_id, note, fatal=fatal, query=query,
  77.             data=json.dumps(data).encode('utf-8') if data else None,
  78.             headers=({'x-viki-app-ver': self._APP_VERSION} if data
  79.                      else self._stream_headers(timestamp, sig) if query is None
  80.                      else None), expected_status=400) or {}
  82.         self._raise_error(resp.get('error'), fatal)
  83.         return resp
  85.     def _raise_error(self, error, fatal=True):
  86.         if error is None:
  87.             return
  88.         msg = '%s said: %s' % (self.IE_NAME, error)
  89.         if fatal:
  90.             raise ExtractorError(msg, expected=True)
  91.         else:
  92.             self.report_warning(msg)
  94.     def _check_errors(self, data):
  95.         for reason, status in (data.get('blocking') or {}).items():
  96.             if status and reason in self._ERRORS:
  97.                 message = self._ERRORS[reason]
  98.                 if reason == 'geo':
  99.                     self.raise_geo_restricted(msg=message)
  100.                 elif reason == 'paywall':
  101.                     if try_get(data, lambda x: x['paywallable']['tvod']):
  102.                         self._raise_error('This video is for rent only or TVOD (Transactional Video On demand)')
  103.                     self.raise_login_required(message)
  104.                 self._raise_error(message)
  106.     def _real_initialize(self):
  107.         self._login()
  109.     def _login(self):
  110.         username, password = self._get_login_info()
  111.         if username is None:
  112.             return
  114.         self._token = self._call_api(
  115.             'sessions.json', None, 'Logging in', fatal=False,
  116.             data={'username': username, 'password': password}).get('token')
  117.         if not self._token:
  118.             self.report_warning('Login Failed: Unable to get session token')
  120.     @staticmethod
  121.     def dict_selection(dict_obj, preferred_key):
  122.         if preferred_key in dict_obj:
  123.             return dict_obj[preferred_key]
  124.         return (list(filter(None, dict_obj.values())) or [None])[0]
  127. class VikiIE(VikiBaseIE):
  128.     IE_NAME = 'viki'
  129.     _VALID_URL = r'%s(?:videos|player)/(?P<id>[0-9]+v)' % VikiBaseIE._VALID_URL_BASE
  130.     _TESTS = [{
  131.         'note': 'Free non-DRM video with storyboards in MPD',
  132.         'url': 'https://www.viki.com/videos/1175236v-choosing-spouse-by-lottery-episode-1',
  133.         'info_dict': {
  134.             'id': '1175236v',
  135.             'ext': 'mp4',
  136.             'title': 'Choosing Spouse by Lottery - Episode 1',
  137.             'timestamp': 1606463239,
  138.             'age_limit': 12,
  139.             'uploader': 'FCC',
  140.             'upload_date': '20201127',
  141.         },
  142.         'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
  143.         'params': {
  144.             'format': 'bestvideo',
  145.         },
  146.     }, {
  147.         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
  148.         'info_dict': {
  149.             'id': '1023585v',
  150.             'ext': 'mp4',
  151.             'title': 'Heirs - Episode 14',
  152.             'uploader': 'SBS Contents Hub',
  153.             'timestamp': 1385047627,
  154.             'upload_date': '20131121',
  155.             'age_limit': 13,
  156.             'duration': 3570,
  157.             'episode_number': 14,
  158.         },
  159.         'params': {
  160.             'format': 'bestvideo',
  161.         },
  162.         'skip': 'Content is only available to Viki Pass Plus subscribers',
  163.         'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
  164.     }, {
  165.         # clip
  166.         'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
  167.         'md5': '86c0b5dbd4d83a6611a79987cc7a1989',
  168.         'info_dict': {
  169.             'id': '1067139v',
  170.             'ext': 'mp4',
  171.             'title': "'The Avengers: Age of Ultron' Press Conference",
  172.             'description': 'md5:d70b2f9428f5488321bfe1db10d612ea',
  173.             'duration': 352,
  174.             'timestamp': 1430380829,
  175.             'upload_date': '20150430',
  176.             'uploader': 'Arirang TV',
  177.             'like_count': int,
  178.             'age_limit': 0,
  179.         },
  180.         'skip': 'Sorry. There was an error loading this video',
  181.     }, {
  182.         'url': 'http://www.viki.com/videos/1048879v-ankhon-dekhi',
  183.         'info_dict': {
  184.             'id': '1048879v',
  185.             'ext': 'mp4',
  186.             'title': 'Ankhon Dekhi',
  187.             'duration': 6512,
  188.             'timestamp': 1408532356,
  189.             'upload_date': '20140820',
  190.             'uploader': 'Spuul',
  191.             'like_count': int,
  192.             'age_limit': 13,
  193.         },
  194.         'skip': 'Page not found!',
  195.     }, {
  196.         # episode
  197.         'url': 'http://www.viki.com/videos/44699v-boys-over-flowers-episode-1',
  198.         'md5': '670440c79f7109ca6564d4c7f24e3e81',
  199.         'info_dict': {
  200.             'id': '44699v',
  201.             'ext': 'mp4',
  202.             'title': 'Boys Over Flowers - Episode 1',
  203.             'description': 'md5:b89cf50038b480b88b5b3c93589a9076',
  204.             'duration': 4172,
  205.             'timestamp': 1270496524,
  206.             'upload_date': '20100405',
  207.             'uploader': 'group8',
  208.             'like_count': int,
  209.             'age_limit': 15,
  210.             'episode_number': 1,
  211.         },
  212.         'params': {
  213.             'format': 'bestvideo',
  214.         },
  215.         'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
  216.     }, {
  217.         # youtube external
  218.         'url': 'http://www.viki.com/videos/50562v-poor-nastya-complete-episode-1',
  219.         'md5': '63f8600c1da6f01b7640eee7eca4f1da',
  220.         'info_dict': {
  221.             'id': '50562v',
  222.             'ext': 'webm',
  223.             'title': 'Poor Nastya [COMPLETE] - Episode 1',
  224.             'description': '',
  225.             'duration': 606,
  226.             'timestamp': 1274949505,
  227.             'upload_date': '20101213',
  228.             'uploader': 'ad14065n',
  229.             'uploader_id': 'ad14065n',
  230.             'like_count': int,
  231.             'age_limit': 13,
  232.         },
  233.         'skip': 'Page not found!',
  234.     }, {
  235.         'url': 'http://www.viki.com/player/44699v',
  236.         'only_matching': True,
  237.     }, {
  238.         # non-English description
  239.         'url': 'http://www.viki.com/videos/158036v-love-in-magic',
  240.         'md5': '78bf49fdaa51f9e7f9150262a9ef9bdf',
  241.         'info_dict': {
  242.             'id': '158036v',
  243.             'ext': 'mp4',
  244.             'uploader': 'I Planet Entertainment',
  245.             'upload_date': '20111122',
  246.             'timestamp': 1321985454,
  247.             'description': 'md5:44b1e46619df3a072294645c770cef36',
  248.             'title': 'Love in Magic',
  249.             'age_limit': 15,
  250.         },
  251.         'params': {
  252.             'format': 'bestvideo',
  253.         },
  254.         'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
  255.     }]
  257.     def _real_extract(self, url):
  258.         video_id = self._match_id(url)
  260.         video = self._call_api('videos/{0}.json'.format(video_id), video_id, 'Downloading video JSON', query={})
  262.         self._check_errors(video)
  264.         title = try_get(video, lambda x: x['titles']['en'], str)
  265.         episode_number = int_or_none(video.get('number'))
  266.         if not title:
  267.             title = 'Episode %d' % episode_number if video.get('type') == 'episode' else video.get('id') or video_id
  268.             container_titles = try_get(video, lambda x: x['container']['titles'], dict) or {}
  269.             container_title = self.dict_selection(container_titles, 'en')
  270.             if container_title and title == video_id:
  271.                 title = container_title
  272.             else:
  273.                 title = '%s - %s' % (container_title, title)
  275.         resp = self._call_api(
  276.             'playback_streams/%s.json?drms=dt3&device_id=%s' % (video_id, self._DEVICE_ID),
  277.             video_id, 'Downloading video streams JSON')['main'][0]
  279.         mpd_url = resp['url']
  280.         # 720p is hidden in another MPD which can be found in the current manifest content
  281.         mpd_content = self._download_webpage(mpd_url, video_id, note='Downloading initial MPD manifest')
  282.         mpd_url = self._search_regex(
  283.             r'(?mi)<BaseURL>(http.+.mpd)', mpd_content, 'new manifest', default=mpd_url)
  284.         if 'mpdhd_high' not in mpd_url:
  285.             # Modify the URL to get 1080p
  286.             mpd_url = mpd_url.replace('mpdhd', 'mpdhd_high')
  287.         formats = self._extract_mpd_formats(mpd_url, video_id)
  288.         self._sort_formats(formats)
  290.         description = self.dict_selection(video.get('descriptions', {}), 'en')
  291.         thumbnails = [{
  292.             'id': thumbnail_id,
  293.             'url': thumbnail['url'],
  294.         } for thumbnail_id, thumbnail in (video.get('images') or {}).items() if thumbnail.get('url')]
  295.         like_count = int_or_none(try_get(video, lambda x: x['likes']['count']))
  297.         stream_id = try_get(resp, lambda x: x['properties']['track']['stream_id'])
  298.         subtitles = dict((lang, [{
  299.             'ext': ext,
  300.             'url': self._API_URL_TEMPLATE % self._api_query(
  301.                 'videos/{0}/auth_subtitles/{1}.{2}'.format(video_id, lang, ext), stream_id=stream_id)
  302.         } for ext in ('srt', 'vtt')]) for lang in (video.get('subtitle_completions') or {}).keys())
  304.         return {
  305.             'id': video_id,
  306.             'formats': formats,
  307.             'title': title,
  308.             'description': description,
  309.             'duration': int_or_none(video.get('duration')),
  310.             'timestamp': parse_iso8601(video.get('created_at')),
  311.             'uploader': video.get('author'),
  312.             'uploader_url': video.get('author_url'),
  313.             'like_count': like_count,
  314.             'age_limit': parse_age_limit(video.get('rating')),
  315.             'thumbnails': thumbnails,
  316.             'subtitles': subtitles,
  317.             'episode_number': episode_number,
  318.         }
  321. class VikiChannelIE(VikiBaseIE):
  322.     IE_NAME = 'viki:channel'
  323.     _VALID_URL = r'%s(?:tv|news|movies|artists)/(?P<id>[0-9]+c)' % VikiBaseIE._VALID_URL_BASE
  324.     _TESTS = [{
  325.         'url': 'http://www.viki.com/tv/50c-boys-over-flowers',
  326.         'info_dict': {
  327.             'id': '50c',
  328.             'title': 'Boys Over Flowers',
  329.             'description': 'md5:f08b679c200e1a273c695fe9986f21d7',
  330.         },
  331.         'playlist_mincount': 51,
  332.     }, {
  333.         'url': 'http://www.viki.com/tv/1354c-poor-nastya-complete',
  334.         'info_dict': {
  335.             'id': '1354c',
  336.             'title': 'Poor Nastya [COMPLETE]',
  337.             'description': 'md5:05bf5471385aa8b21c18ad450e350525',
  338.         },
  339.         'playlist_count': 127,
  340.         'skip': 'Page not found',
  341.     }, {
  342.         'url': 'http://www.viki.com/news/24569c-showbiz-korea',
  343.         'only_matching': True,
  344.     }, {
  345.         'url': 'http://www.viki.com/movies/22047c-pride-and-prejudice-2005',
  346.         'only_matching': True,
  347.     }, {
  348.         'url': 'http://www.viki.com/artists/2141c-shinee',
  349.         'only_matching': True,
  350.     }]
  352.     _video_types = ('episodes', 'movies', 'clips', 'trailers')
  354.     def _entries(self, channel_id):
  355.         params = {
  356.             'app': self._APP, 'token': self._token, 'only_ids': 'true',
  357.             'direction': 'asc', 'sort': 'number', 'per_page': 30
  358.         }
  359.         video_types = self._video_types
  360.         for video_type in video_types:
  361.             if video_type not in self._video_types:
  362.                 self.report_warning('Unknown video_type: ' + video_type)
  363.             page_num = 0
  364.             while True:
  365.                 page_num += 1
  366.                 params['page'] = page_num
  367.                 res = self._call_api(
  368.                     'containers/{channel_id}/{video_type}.json'.format(**locals()), channel_id, query=params, fatal=False,
  369.                     note='Downloading %s JSON page %d' % (video_type.title(), page_num))
  371.                 for video_id in res.get('response') or []:
  372.                     yield self.url_result('https://www.viki.com/videos/' + video_id, VikiIE.ie_key(), video_id)
  373.                 if not res.get('more'):
  374.                     break
  376.     def _real_extract(self, url):
  377.         channel_id = self._match_id(url)
  379.         channel = self._call_api('containers/%s.json' % channel_id, channel_id, 'Downloading channel JSON')
  381.         self._check_errors(channel)
  383.         return self.playlist_result(
  384.             self._entries(channel_id), channel_id,
  385.             self.dict_selection(channel['titles'], 'en'),
  386.             self.dict_selection(channel['descriptions'], 'en'))