Raphaël G. Git Repositories - youtubedl/blob - 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             index = self._call_api('index', video_id, 'CMS Signing')
  68             self._CMS_SIGNING = index.get('cms_signing') or {}
  69             if not self._CMS_SIGNING:
  70                 for signing_policy in index.get('signing_policies', []):
  71                     signing_path = signing_policy.get('path')
  72                     if signing_path and signing_path.startswith('/cms/'):
  73                         name, value = signing_policy.get('name'), signing_policy.get('value')
  74                         if name and value:
  75                             self._CMS_SIGNING[name] = value
  76         return self._download_json(
  77             self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
  78             note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
  79
  80     def _get_cms_resource(self, resource_key, video_id):
  81         return self._call_api(
  82             'cms_resource', video_id, 'resource path', data={
  83                 'resource_key': resource_key,
  84             })['__links__']['cms_resource']['href']
  85
  86     def _real_initialize(self):
  87         webpage = self._download_webpage(
  88             'https://vrv.co/', None, headers=self.geo_verification_headers())
  89         self._API_PARAMS = self._parse_json(self._search_regex(
  90             [
  91                 r'window\.__APP_CONFIG__\s*=\s*({.+?})(?:</script>|;)',
  92                 r'window\.__APP_CONFIG__\s*=\s*({.+})'
  93             ], webpage, 'app config'), None)['cxApiParams']
  94         self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
  95
  96
  97 class VRVIE(VRVBaseIE):
  98     IE_NAME = 'vrv'
  99     _VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
 100     _TESTS = [{
 101         'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
 102         'info_dict': {
 103             'id': 'GR9PNZ396',
 104             'ext': 'mp4',
 105             'title': 'BOSTON: WHERE THE PAST IS THE PRESENT',
 106             'description': 'md5:4ec8844ac262ca2df9e67c0983c6b83f',
 107             'uploader_id': 'seeso',
 108         },
 109         'params': {
 110             # m3u8 download
 111             'skip_download': True,
 112         },
 113     }, {
 114         # movie listing
 115         'url': 'https://vrv.co/watch/G6NQXZ1J6/Lily-CAT',
 116         'info_dict': {
 117             'id': 'G6NQXZ1J6',
 118             'title': 'Lily C.A.T',
 119             'description': 'md5:988b031e7809a6aeb60968be4af7db07',
 120         },
 121         'playlist_count': 2,
 122     }]
 123     _NETRC_MACHINE = 'vrv'
 124
 125     def _real_initialize(self):
 126         super(VRVIE, self)._real_initialize()
 127
 128         email, password = self._get_login_info()
 129         if email is None:
 130             return
 131
 132         token_credentials = self._call_api(
 133             'authenticate/by:credentials', None, 'Token Credentials', data={
 134                 'email': email,
 135                 'password': password,
 136             })
 137         self._TOKEN = token_credentials['oauth_token']
 138         self._TOKEN_SECRET = token_credentials['oauth_token_secret']
 139
 140     def _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
 141         if not url or stream_format not in ('hls', 'dash', 'adaptive_hls'):
 142             return []
 143         stream_id_list = []
 144         if audio_lang:
 145             stream_id_list.append('audio-%s' % audio_lang)
 146         if hardsub_lang:
 147             stream_id_list.append('hardsub-%s' % hardsub_lang)
 148         format_id = stream_format
 149         if stream_id_list:
 150             format_id += '-' + '-'.join(stream_id_list)
 151         if 'hls' in stream_format:
 152             adaptive_formats = self._extract_m3u8_formats(
 153                 url, video_id, 'mp4', m3u8_id=format_id,
 154                 note='Downloading %s information' % format_id,
 155                 fatal=False)
 156         elif stream_format == 'dash':
 157             adaptive_formats = self._extract_mpd_formats(
 158                 url, video_id, mpd_id=format_id,
 159                 note='Downloading %s information' % format_id,
 160                 fatal=False)
 161         if audio_lang:
 162             for f in adaptive_formats:
 163                 if f.get('acodec') != 'none':
 164                     f['language'] = audio_lang
 165         return adaptive_formats
 166
 167     def _real_extract(self, url):
 168         video_id = self._match_id(url)
 169
 170         object_data = self._call_cms(self._get_cms_resource(
 171             'cms:/objects/' + video_id, video_id), video_id, 'object')['items'][0]
 172         resource_path = object_data['__links__']['resource']['href']
 173         video_data = self._call_cms(resource_path, video_id, 'video')
 174         title = video_data['title']
 175         description = video_data.get('description')
 176
 177         if video_data.get('__class__') == 'movie_listing':
 178             items = self._call_cms(
 179                 video_data['__links__']['movie_listing/movies']['href'],
 180                 video_id, 'movie listing').get('items') or []
 181             if len(items) != 1:
 182                 entries = []
 183                 for item in items:
 184                     item_id = item.get('id')
 185                     if not item_id:
 186                         continue
 187                     entries.append(self.url_result(
 188                         'https://vrv.co/watch/' + item_id,
 189                         self.ie_key(), item_id, item.get('title')))
 190                 return self.playlist_result(entries, video_id, title, description)
 191             video_data = items[0]
 192
 193         streams_path = video_data['__links__'].get('streams', {}).get('href')
 194         if not streams_path:
 195             self.raise_login_required()
 196         streams_json = self._call_cms(streams_path, video_id, 'streams')
 197
 198         audio_locale = streams_json.get('audio_locale')
 199         formats = []
 200         for stream_type, streams in streams_json.get('streams', {}).items():
 201             if stream_type in ('adaptive_hls', 'adaptive_dash'):
 202                 for stream in streams.values():
 203                     formats.extend(self._extract_vrv_formats(
 204                         stream.get('url'), video_id, stream_type.split('_')[1],
 205                         audio_locale, stream.get('hardsub_locale')))
 206         self._sort_formats(formats)
 207
 208         subtitles = {}
 209         for k in ('captions', 'subtitles'):
 210             for subtitle in streams_json.get(k, {}).values():
 211                 subtitle_url = subtitle.get('url')
 212                 if not subtitle_url:
 213                     continue
 214                 subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
 215                     'url': subtitle_url,
 216                     'ext': subtitle.get('format', 'ass'),
 217                 })
 218
 219         thumbnails = []
 220         for thumbnail in video_data.get('images', {}).get('thumbnails', []):
 221             thumbnail_url = thumbnail.get('source')
 222             if not thumbnail_url:
 223                 continue
 224             thumbnails.append({
 225                 'url': thumbnail_url,
 226                 'width': int_or_none(thumbnail.get('width')),
 227                 'height': int_or_none(thumbnail.get('height')),
 228             })
 229
 230         return {
 231             'id': video_id,
 232             'title': title,
 233             'formats': formats,
 234             'subtitles': subtitles,
 235             'thumbnails': thumbnails,
 236             'description': description,
 237             'duration': float_or_none(video_data.get('duration_ms'), 1000),
 238             'uploader_id': video_data.get('channel_id'),
 239             'series': video_data.get('series_title'),
 240             'season': video_data.get('season_title'),
 241             'season_number': int_or_none(video_data.get('season_number')),
 242             'season_id': video_data.get('season_id'),
 243             'episode': title,
 244             'episode_number': int_or_none(video_data.get('episode_number')),
 245             'episode_id': video_data.get('production_episode_id'),
 246         }
 247
 248
 249 class VRVSeriesIE(VRVBaseIE):
 250     IE_NAME = 'vrv:series'
 251     _VALID_URL = r'https?://(?:www\.)?vrv\.co/series/(?P<id>[A-Z0-9]+)'
 252     _TEST = {
 253         'url': 'https://vrv.co/series/G68VXG3G6/The-Perfect-Insider',
 254         'info_dict': {
 255             'id': 'G68VXG3G6',
 256         },
 257         'playlist_mincount': 11,
 258     }
 259
 260     def _real_extract(self, url):
 261         series_id = self._match_id(url)
 262
 263         seasons_path = self._get_cms_resource(
 264             'cms:/seasons?series_id=' + series_id, series_id)
 265         seasons_data = self._call_cms(seasons_path, series_id, 'seasons')
 266
 267         entries = []
 268         for season in seasons_data.get('items', []):
 269             episodes_path = season['__links__']['season/episodes']['href']
 270             episodes = self._call_cms(episodes_path, series_id, 'episodes')
 271             for episode in episodes.get('items', []):
 272                 episode_id = episode['id']
 273                 entries.append(self.url_result(
 274                     'https://vrv.co/watch/' + episode_id,
 275                     'VRV', episode_id, episode.get('title')))
 276
 277         return self.playlist_result(entries, series_id)