youtube_dl/extractor/vrv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import json
   6 import hashlib
   7 import hmac
   8 import random
   9 import string
  10 import time
  11
  12 from .common import InfoExtractor
  13 from ..compat import (
  14     compat_HTTPError,
  15     compat_urllib_parse_urlencode,
  16     compat_urllib_parse,
  17 )
  18 from ..utils import (
  19     ExtractorError,
  20     float_or_none,
  21     int_or_none,
  22 )
  23
  24
  25 class VRVBaseIE(InfoExtractor):
  26     _API_DOMAIN = None
  27     _API_PARAMS = {}
  28     _CMS_SIGNING = {}
  29     _TOKEN = None
  30     _TOKEN_SECRET = ''
  31
  32     def _call_api(self, path, video_id, note, data=None):
  33         # https://tools.ietf.org/html/rfc5849#section-3
  34         base_url = self._API_DOMAIN + '/core/' + path
  35         query = [
  36             ('oauth_consumer_key', self._API_PARAMS['oAuthKey']),
  37             ('oauth_nonce', ''.join([random.choice(string.ascii_letters) for _ in range(32)])),
  38             ('oauth_signature_method', 'HMAC-SHA1'),
  39             ('oauth_timestamp', int(time.time())),
  40         ]
  41         if self._TOKEN:
  42             query.append(('oauth_token', self._TOKEN))
  43         encoded_query = compat_urllib_parse_urlencode(query)
  44         headers = self.geo_verification_headers()
  45         if data:
  46             data = json.dumps(data).encode()
  47             headers['Content-Type'] = 'application/json'
  48         base_string = '&'.join([
  49             'POST' if data else 'GET',
  50             compat_urllib_parse.quote(base_url, ''),
  51             compat_urllib_parse.quote(encoded_query, '')])
  52         oauth_signature = base64.b64encode(hmac.new(
  53             (self._API_PARAMS['oAuthSecret'] + '&' + self._TOKEN_SECRET).encode('ascii'),
  54             base_string.encode(), hashlib.sha1).digest()).decode()
  55         encoded_query += '&oauth_signature=' + compat_urllib_parse.quote(oauth_signature, '')
  56         try:
  57             return self._download_json(
  58                 '?'.join([base_url, encoded_query]), video_id,
  59                 note='Downloading %s JSON metadata' % note, headers=headers, data=data)
  60         except ExtractorError as e:
  61             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
  62                 raise ExtractorError(json.loads(e.cause.read().decode())['message'], expected=True)
  63             raise
  64
  65     def _call_cms(self, path, video_id, note):
  66         if not self._CMS_SIGNING:
  67             self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing']
  68         return self._download_json(
  69             self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
  70             note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
  71
  72     def _get_cms_resource(self, resource_key, video_id):
  73         return self._call_api(
  74             'cms_resource', video_id, 'resource path', data={
  75                 'resource_key': resource_key,
  76             })['__links__']['cms_resource']['href']
  77
  78     def _real_initialize(self):
  79         webpage = self._download_webpage(
  80             'https://vrv.co/', None, headers=self.geo_verification_headers())
  81         self._API_PARAMS = self._parse_json(self._search_regex(
  82             [
  83                 r'window\.__APP_CONFIG__\s*=\s*({.+?})(?:</script>|;)',
  84                 r'window\.__APP_CONFIG__\s*=\s*({.+})'
  85             ], webpage, 'app config'), None)['cxApiParams']
  86         self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
  87
  88
  89 class VRVIE(VRVBaseIE):
  90     IE_NAME = 'vrv'
  91     _VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
  92     _TESTS = [{
  93         'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
  94         'info_dict': {
  95             'id': 'GR9PNZ396',
  96             'ext': 'mp4',
  97             'title': 'BOSTON: WHERE THE PAST IS THE PRESENT',
  98             'description': 'md5:4ec8844ac262ca2df9e67c0983c6b83f',
  99             'uploader_id': 'seeso',
 100         },
 101         'params': {
 102             # m3u8 download
 103             'skip_download': True,
 104         },
 105     }]
 106     _NETRC_MACHINE = 'vrv'
 107
 108     def _real_initialize(self):
 109         super(VRVIE, self)._real_initialize()
 110
 111         email, password = self._get_login_info()
 112         if email is None:
 113             return
 114
 115         token_credentials = self._call_api(
 116             'authenticate/by:credentials', None, 'Token Credentials', data={
 117                 'email': email,
 118                 'password': password,
 119             })
 120         self._TOKEN = token_credentials['oauth_token']
 121         self._TOKEN_SECRET = token_credentials['oauth_token_secret']
 122
 123     def _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
 124         if not url or stream_format not in ('hls', 'dash'):
 125             return []
 126         assert audio_lang or hardsub_lang
 127         stream_id_list = []
 128         if audio_lang:
 129             stream_id_list.append('audio-%s' % audio_lang)
 130         if hardsub_lang:
 131             stream_id_list.append('hardsub-%s' % hardsub_lang)
 132         stream_id = '-'.join(stream_id_list)
 133         format_id = '%s-%s' % (stream_format, stream_id)
 134         if stream_format == 'hls':
 135             adaptive_formats = self._extract_m3u8_formats(
 136                 url, video_id, 'mp4', m3u8_id=format_id,
 137                 note='Downloading %s m3u8 information' % stream_id,
 138                 fatal=False)
 139         elif stream_format == 'dash':
 140             adaptive_formats = self._extract_mpd_formats(
 141                 url, video_id, mpd_id=format_id,
 142                 note='Downloading %s MPD information' % stream_id,
 143                 fatal=False)
 144         if audio_lang:
 145             for f in adaptive_formats:
 146                 if f.get('acodec') != 'none':
 147                     f['language'] = audio_lang
 148         return adaptive_formats
 149
 150     def _real_extract(self, url):
 151         video_id = self._match_id(url)
 152
 153         object_data = self._call_cms(self._get_cms_resource(
 154             'cms:/objects/' + video_id, video_id), video_id, 'object')['items'][0]
 155         resource_path = object_data['__links__']['resource']['href']
 156         video_data = self._call_cms(resource_path, video_id, 'video')
 157         title = video_data['title']
 158
 159         streams_path = video_data['__links__'].get('streams', {}).get('href')
 160         if not streams_path:
 161             self.raise_login_required()
 162         streams_json = self._call_cms(streams_path, video_id, 'streams')
 163
 164         audio_locale = streams_json.get('audio_locale')
 165         formats = []
 166         for stream_type, streams in streams_json.get('streams', {}).items():
 167             if stream_type in ('adaptive_hls', 'adaptive_dash'):
 168                 for stream in streams.values():
 169                     formats.extend(self._extract_vrv_formats(
 170                         stream.get('url'), video_id, stream_type.split('_')[1],
 171                         audio_locale, stream.get('hardsub_locale')))
 172         self._sort_formats(formats)
 173
 174         subtitles = {}
 175         for subtitle in streams_json.get('subtitles', {}).values():
 176             subtitle_url = subtitle.get('url')
 177             if not subtitle_url:
 178                 continue
 179             subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
 180                 'url': subtitle_url,
 181                 'ext': subtitle.get('format', 'ass'),
 182             })
 183
 184         thumbnails = []
 185         for thumbnail in video_data.get('images', {}).get('thumbnails', []):
 186             thumbnail_url = thumbnail.get('source')
 187             if not thumbnail_url:
 188                 continue
 189             thumbnails.append({
 190                 'url': thumbnail_url,
 191                 'width': int_or_none(thumbnail.get('width')),
 192                 'height': int_or_none(thumbnail.get('height')),
 193             })
 194
 195         return {
 196             'id': video_id,
 197             'title': title,
 198             'formats': formats,
 199             'subtitles': subtitles,
 200             'thumbnails': thumbnails,
 201             'description': video_data.get('description'),
 202             'duration': float_or_none(video_data.get('duration_ms'), 1000),
 203             'uploader_id': video_data.get('channel_id'),
 204             'series': video_data.get('series_title'),
 205             'season': video_data.get('season_title'),
 206             'season_number': int_or_none(video_data.get('season_number')),
 207             'season_id': video_data.get('season_id'),
 208             'episode': title,
 209             'episode_number': int_or_none(video_data.get('episode_number')),
 210             'episode_id': video_data.get('production_episode_id'),
 211         }
 212
 213
 214 class VRVSeriesIE(VRVBaseIE):
 215     IE_NAME = 'vrv:series'
 216     _VALID_URL = r'https?://(?:www\.)?vrv\.co/series/(?P<id>[A-Z0-9]+)'
 217     _TEST = {
 218         'url': 'https://vrv.co/series/G68VXG3G6/The-Perfect-Insider',
 219         'info_dict': {
 220             'id': 'G68VXG3G6',
 221         },
 222         'playlist_mincount': 11,
 223     }
 224
 225     def _real_extract(self, url):
 226         series_id = self._match_id(url)
 227
 228         seasons_path = self._get_cms_resource(
 229             'cms:/seasons?series_id=' + series_id, series_id)
 230         seasons_data = self._call_cms(seasons_path, series_id, 'seasons')
 231
 232         entries = []
 233         for season in seasons_data.get('items', []):
 234             episodes_path = season['__links__']['season/episodes']['href']
 235             episodes = self._call_cms(episodes_path, series_id, 'episodes')
 236             for episode in episodes.get('items', []):
 237                 episode_id = episode['id']
 238                 entries.append(self.url_result(
 239                     'https://vrv.co/watch/' + episode_id,
 240                     'VRV', episode_id, episode.get('title')))
 241
 242         return self.playlist_result(entries, series_id)