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_urlparse,
  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_urlparse.quote(base_url, ''), compat_urlparse.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_urlparse.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     _TEST = {
  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 _real_extract(self, url):
  91         video_id = self._match_id(url)
  92         webpage = self._download_webpage(
  93             url, video_id,
  94             headers=self.geo_verification_headers())
  95         media_resource = self._parse_json(self._search_regex(
  96             r'window\.__INITIAL_STATE__\s*=\s*({.+?})</script>',
  97             webpage, 'inital state'), video_id).get('watch', {}).get('mediaResource') or {}
  98
  99         video_data = media_resource.get('json')
 100         if not video_data:
 101             self._set_api_params(webpage, video_id)
 102             episode_path = self._get_cms_resource(
 103                 'cms:/episodes/' + video_id, video_id)
 104             video_data = self._call_cms(episode_path, video_id, 'video')
 105         title = video_data['title']
 106
 107         streams_json = media_resource.get('streams', {}).get('json', {})
 108         if not streams_json:
 109             self._set_api_params(webpage, video_id)
 110             streams_path = video_data['__links__']['streams']['href']
 111             streams_json = self._call_cms(streams_path, video_id, 'streams')
 112
 113         audio_locale = streams_json.get('audio_locale')
 114         formats = []
 115         for stream_id, stream in streams_json.get('streams', {}).get('adaptive_hls', {}).items():
 116             stream_url = stream.get('url')
 117             if not stream_url:
 118                 continue
 119             stream_id = stream_id or audio_locale
 120             m3u8_formats = self._extract_m3u8_formats(
 121                 stream_url, video_id, 'mp4', m3u8_id=stream_id,
 122                 note='Downloading %s m3u8 information' % stream_id,
 123                 fatal=False)
 124             if audio_locale:
 125                 for f in m3u8_formats:
 126                     f['language'] = audio_locale
 127             formats.extend(m3u8_formats)
 128         self._sort_formats(formats)
 129
 130         thumbnails = []
 131         for thumbnail in video_data.get('images', {}).get('thumbnails', []):
 132             thumbnail_url = thumbnail.get('source')
 133             if not thumbnail_url:
 134                 continue
 135             thumbnails.append({
 136                 'url': thumbnail_url,
 137                 'width': int_or_none(thumbnail.get('width')),
 138                 'height': int_or_none(thumbnail.get('height')),
 139             })
 140
 141         return {
 142             'id': video_id,
 143             'title': title,
 144             'formats': formats,
 145             'thumbnails': thumbnails,
 146             'description': video_data.get('description'),
 147             'duration': float_or_none(video_data.get('duration_ms'), 1000),
 148             'uploader_id': video_data.get('channel_id'),
 149             'series': video_data.get('series_title'),
 150             'season': video_data.get('season_title'),
 151             'season_number': int_or_none(video_data.get('season_number')),
 152             'season_id': video_data.get('season_id'),
 153             'episode': title,
 154             'episode_number': int_or_none(video_data.get('episode_number')),
 155             'episode_id': video_data.get('production_episode_id'),
 156         }
 157
 158
 159 class VRVSeriesIE(VRVBaseIE):
 160     IE_NAME = 'vrv:series'
 161     _VALID_URL = r'https?://(?:www\.)?vrv\.co/series/(?P<id>[A-Z0-9]+)'
 162     _TEST = {
 163         'url': 'https://vrv.co/series/G68VXG3G6/The-Perfect-Insider',
 164         'info_dict': {
 165             'id': 'G68VXG3G6',
 166         },
 167         'playlist_mincount': 11,
 168     }
 169
 170     def _real_extract(self, url):
 171         series_id = self._match_id(url)
 172         webpage = self._download_webpage(
 173             url, series_id,
 174             headers=self.geo_verification_headers())
 175
 176         self._set_api_params(webpage, series_id)
 177         seasons_path = self._get_cms_resource(
 178             'cms:/seasons?series_id=' + series_id, series_id)
 179         seasons_data = self._call_cms(seasons_path, series_id, 'seasons')
 180
 181         entries = []
 182         for season in seasons_data.get('items', []):
 183             episodes_path = season['__links__']['season/episodes']['href']
 184             episodes = self._call_cms(episodes_path, series_id, 'episodes')
 185             for episode in episodes.get('items', []):
 186                 episode_id = episode['id']
 187                 entries.append(self.url_result(
 188                     'https://vrv.co/watch/' + episode_id,
 189                     'VRV', episode_id, episode.get('title')))
 190
 191         return self.playlist_result(entries, series_id)