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_urllib_parse_urlencode,
  15     compat_urllib_parse,
  16 )
  17 from ..utils import (
  18     float_or_none,
  19     int_or_none,
  20 )
  21
  22
  23 class VRVBaseIE(InfoExtractor):
  24     _API_DOMAIN = None
  25     _API_PARAMS = {}
  26     _CMS_SIGNING = {}
  27
  28     def _call_api(self, path, video_id, note, data=None):
  29         base_url = self._API_DOMAIN + '/core/' + path
  30         encoded_query = compat_urllib_parse_urlencode({
  31             'oauth_consumer_key': self._API_PARAMS['oAuthKey'],
  32             'oauth_nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
  33             'oauth_signature_method': 'HMAC-SHA1',
  34             'oauth_timestamp': int(time.time()),
  35             'oauth_version': '1.0',
  36         })
  37         headers = self.geo_verification_headers()
  38         if data:
  39             data = json.dumps(data).encode()
  40             headers['Content-Type'] = 'application/json'
  41         method = 'POST' if data else 'GET'
  42         base_string = '&'.join([method, compat_urllib_parse.quote(base_url, ''), compat_urllib_parse.quote(encoded_query, '')])
  43         oauth_signature = base64.b64encode(hmac.new(
  44             (self._API_PARAMS['oAuthSecret'] + '&').encode('ascii'),
  45             base_string.encode(), hashlib.sha1).digest()).decode()
  46         encoded_query += '&oauth_signature=' + compat_urllib_parse.quote(oauth_signature, '')
  47         return self._download_json(
  48             '?'.join([base_url, encoded_query]), video_id,
  49             note='Downloading %s JSON metadata' % note, headers=headers, data=data)
  50
  51     def _call_cms(self, path, video_id, note):
  52         if not self._CMS_SIGNING:
  53             self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing']
  54         return self._download_json(
  55             self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
  56             note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
  57
  58     def _set_api_params(self, webpage, video_id):
  59         if not self._API_PARAMS:
  60             self._API_PARAMS = self._parse_json(self._search_regex(
  61                 r'window\.__APP_CONFIG__\s*=\s*({.+?})</script>',
  62                 webpage, 'api config'), video_id)['cxApiParams']
  63             self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
  64
  65     def _get_cms_resource(self, resource_key, video_id):
  66         return self._call_api(
  67             'cms_resource', video_id, 'resource path', data={
  68                 'resource_key': resource_key,
  69             })['__links__']['cms_resource']['href']
  70
  71
  72 class VRVIE(VRVBaseIE):
  73     IE_NAME = 'vrv'
  74     _VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
  75     _TESTS = [{
  76         'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
  77         'info_dict': {
  78             'id': 'GR9PNZ396',
  79             'ext': 'mp4',
  80             'title': 'BOSTON: WHERE THE PAST IS THE PRESENT',
  81             'description': 'md5:4ec8844ac262ca2df9e67c0983c6b83f',
  82             'uploader_id': 'seeso',
  83         },
  84         'params': {
  85             # m3u8 download
  86             'skip_download': True,
  87         },
  88     }]
  89
  90     def _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
  91         if not url or stream_format not in ('hls', 'dash'):
  92             return []
  93         assert audio_lang or hardsub_lang
  94         stream_id_list = []
  95         if audio_lang:
  96             stream_id_list.append('audio-%s' % audio_lang)
  97         if hardsub_lang:
  98             stream_id_list.append('hardsub-%s' % hardsub_lang)
  99         stream_id = '-'.join(stream_id_list)
 100         format_id = '%s-%s' % (stream_format, stream_id)
 101         if stream_format == 'hls':
 102             adaptive_formats = self._extract_m3u8_formats(
 103                 url, video_id, 'mp4', m3u8_id=format_id,
 104                 note='Downloading %s m3u8 information' % stream_id,
 105                 fatal=False)
 106         elif stream_format == 'dash':
 107             adaptive_formats = self._extract_mpd_formats(
 108                 url, video_id, mpd_id=format_id,
 109                 note='Downloading %s MPD information' % stream_id,
 110                 fatal=False)
 111         if audio_lang:
 112             for f in adaptive_formats:
 113                 if f.get('acodec') != 'none':
 114                     f['language'] = audio_lang
 115         return adaptive_formats
 116
 117     def _real_extract(self, url):
 118         video_id = self._match_id(url)
 119         webpage = self._download_webpage(
 120             url, video_id,
 121             headers=self.geo_verification_headers())
 122         media_resource = self._parse_json(self._search_regex(
 123             [
 124                 r'window\.__INITIAL_STATE__\s*=\s*({.+?})(?:</script>|;)',
 125                 r'window\.__INITIAL_STATE__\s*=\s*({.+})'
 126             ], webpage, 'inital state'), video_id).get('watch', {}).get('mediaResource') or {}
 127
 128         video_data = media_resource.get('json')
 129         if not video_data:
 130             self._set_api_params(webpage, video_id)
 131             episode_path = self._get_cms_resource(
 132                 'cms:/episodes/' + video_id, video_id)
 133             video_data = self._call_cms(episode_path, video_id, 'video')
 134         title = video_data['title']
 135
 136         streams_json = media_resource.get('streams', {}).get('json', {})
 137         if not streams_json:
 138             self._set_api_params(webpage, video_id)
 139             streams_path = video_data['__links__']['streams']['href']
 140             streams_json = self._call_cms(streams_path, video_id, 'streams')
 141
 142         audio_locale = streams_json.get('audio_locale')
 143         formats = []
 144         for stream_type, streams in streams_json.get('streams', {}).items():
 145             if stream_type in ('adaptive_hls', 'adaptive_dash'):
 146                 for stream in streams.values():
 147                     formats.extend(self._extract_vrv_formats(
 148                         stream.get('url'), video_id, stream_type.split('_')[1],
 149                         audio_locale, stream.get('hardsub_locale')))
 150         self._sort_formats(formats)
 151
 152         subtitles = {}
 153         for subtitle in streams_json.get('subtitles', {}).values():
 154             subtitle_url = subtitle.get('url')
 155             if not subtitle_url:
 156                 continue
 157             subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
 158                 'url': subtitle_url,
 159                 'ext': subtitle.get('format', 'ass'),
 160             })
 161
 162         thumbnails = []
 163         for thumbnail in video_data.get('images', {}).get('thumbnails', []):
 164             thumbnail_url = thumbnail.get('source')
 165             if not thumbnail_url:
 166                 continue
 167             thumbnails.append({
 168                 'url': thumbnail_url,
 169                 'width': int_or_none(thumbnail.get('width')),
 170                 'height': int_or_none(thumbnail.get('height')),
 171             })
 172
 173         return {
 174             'id': video_id,
 175             'title': title,
 176             'formats': formats,
 177             'subtitles': subtitles,
 178             'thumbnails': thumbnails,
 179             'description': video_data.get('description'),
 180             'duration': float_or_none(video_data.get('duration_ms'), 1000),
 181             'uploader_id': video_data.get('channel_id'),
 182             'series': video_data.get('series_title'),
 183             'season': video_data.get('season_title'),
 184             'season_number': int_or_none(video_data.get('season_number')),
 185             'season_id': video_data.get('season_id'),
 186             'episode': title,
 187             'episode_number': int_or_none(video_data.get('episode_number')),
 188             'episode_id': video_data.get('production_episode_id'),
 189         }
 190
 191
 192 class VRVSeriesIE(VRVBaseIE):
 193     IE_NAME = 'vrv:series'
 194     _VALID_URL = r'https?://(?:www\.)?vrv\.co/series/(?P<id>[A-Z0-9]+)'
 195     _TEST = {
 196         'url': 'https://vrv.co/series/G68VXG3G6/The-Perfect-Insider',
 197         'info_dict': {
 198             'id': 'G68VXG3G6',
 199         },
 200         'playlist_mincount': 11,
 201     }
 202
 203     def _real_extract(self, url):
 204         series_id = self._match_id(url)
 205         webpage = self._download_webpage(
 206             url, series_id,
 207             headers=self.geo_verification_headers())
 208
 209         self._set_api_params(webpage, series_id)
 210         seasons_path = self._get_cms_resource(
 211             'cms:/seasons?series_id=' + series_id, series_id)
 212         seasons_data = self._call_cms(seasons_path, series_id, 'seasons')
 213
 214         entries = []
 215         for season in seasons_data.get('items', []):
 216             episodes_path = season['__links__']['season/episodes']['href']
 217             episodes = self._call_cms(episodes_path, series_id, 'episodes')
 218             for episode in episodes.get('items', []):
 219                 episode_id = episode['id']
 220                 entries.append(self.url_result(
 221                     'https://vrv.co/watch/' + episode_id,
 222                     'VRV', episode_id, episode.get('title')))
 223
 224         return self.playlist_result(entries, series_id)