_ Git - youtube-dl/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_urlparse,
  16 )
  17 from ..utils import (
  18     float_or_none,
  19     int_or_none,
  20 )
  21
  22
  23 class VRVIE(InfoExtractor):
  24     _VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
  25     _TEST = {
  26         'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
  27         'info_dict': {
  28             'id': 'GR9PNZ396',
  29             'ext': 'mp4',
  30             'title': 'BOSTON: WHERE THE PAST IS THE PRESENT',
  31             'description': 'md5:4ec8844ac262ca2df9e67c0983c6b83f',
  32             'uploader_id': 'seeso',
  33         },
  34         'params': {
  35             # m3u8 download
  36             'skip_download': True,
  37         },
  38     }
  39     _API_DOMAIN = None
  40     _API_PARAMS = {}
  41     _CMS_SIGNING = {}
  42
  43     def _call_api(self, path, video_id, note, data=None):
  44         base_url = self._API_DOMAIN + '/core/' + path
  45         encoded_query = compat_urllib_parse_urlencode({
  46             'oauth_consumer_key': self._API_PARAMS['oAuthKey'],
  47             'oauth_nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
  48             'oauth_signature_method': 'HMAC-SHA1',
  49             'oauth_timestamp': int(time.time()),
  50             'oauth_version': '1.0',
  51         })
  52         headers = self.geo_verification_headers()
  53         if data:
  54             data = json.dumps(data).encode()
  55             headers['Content-Type'] = 'application/json'
  56         method = 'POST' if data else 'GET'
  57         base_string = '&'.join([method, compat_urlparse.quote(base_url, ''), compat_urlparse.quote(encoded_query, '')])
  58         oauth_signature = base64.b64encode(hmac.new(
  59             (self._API_PARAMS['oAuthSecret'] + '&').encode('ascii'),
  60             base_string.encode(), hashlib.sha1).digest()).decode()
  61         encoded_query += '&oauth_signature=' + compat_urlparse.quote(oauth_signature, '')
  62         return self._download_json(
  63             '?'.join([base_url, encoded_query]), video_id,
  64             note='Downloading %s JSON metadata' % note, headers=headers, data=data)
  65
  66     def _call_cms(self, path, video_id, note):
  67         return self._download_json(
  68             self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
  69             note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
  70
  71     def _set_api_params(self, webpage, video_id):
  72         if not self._API_PARAMS:
  73             self._API_PARAMS = self._parse_json(self._search_regex(
  74                 r'window\.__APP_CONFIG__\s*=\s*({.+?})</script>',
  75                 webpage, 'api config'), video_id)['cxApiParams']
  76             self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
  77
  78     def _set_cms_signing(self, video_id):
  79         if not self._CMS_SIGNING:
  80             self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing']
  81
  82     def _real_extract(self, url):
  83         video_id = self._match_id(url)
  84         webpage = self._download_webpage(
  85             url, video_id,
  86             headers=self.geo_verification_headers())
  87         media_resource = self._parse_json(self._search_regex(
  88             r'window\.__INITIAL_STATE__\s*=\s*({.+?})</script>',
  89             webpage, 'inital state'), video_id).get('watch', {}).get('mediaResource') or {}
  90
  91         video_data = media_resource.get('json')
  92         if not video_data:
  93             self._set_api_params(webpage, video_id)
  94             episode_path = self._call_api('cms_resource', video_id, 'episode resource path', data={
  95                 'resource_key': 'cms:/episodes/' + video_id,
  96             })['__links__']['cms_resource']['href']
  97             self._set_cms_signing(video_id)
  98             video_data = self._call_cms(episode_path, video_id, 'video')
  99         title = video_data['title']
 100
 101         streams_json = media_resource.get('streams', {}).get('json', {})
 102         if not streams_json:
 103             self._set_api_params(webpage, video_id)
 104             streams_path = video_data['__links__']['streams']['href']
 105             self._set_cms_signing(video_id)
 106             streams_json = self._call_cms(streams_path, video_id, 'streams')
 107
 108         audio_locale = streams_json.get('audio_locale')
 109         formats = []
 110         for stream_id, stream in streams_json.get('streams', {}).get('adaptive_hls', {}).items():
 111             stream_url = stream.get('url')
 112             if not stream_url:
 113                 continue
 114             stream_id = stream_id or audio_locale
 115             m3u8_formats = self._extract_m3u8_formats(
 116                 stream_url, video_id, 'mp4', m3u8_id=stream_id,
 117                 note='Downloading %s m3u8 information' % stream_id,
 118                 fatal=False)
 119             if audio_locale:
 120                 for f in m3u8_formats:
 121                     f['language'] = audio_locale
 122             formats.extend(m3u8_formats)
 123         self._sort_formats(formats)
 124
 125         thumbnails = []
 126         for thumbnail in video_data.get('images', {}).get('thumbnails', []):
 127             thumbnail_url = thumbnail.get('source')
 128             if not thumbnail_url:
 129                 continue
 130             thumbnails.append({
 131                 'url': thumbnail_url,
 132                 'width': int_or_none(thumbnail.get('width')),
 133                 'height': int_or_none(thumbnail.get('height')),
 134             })
 135
 136         return {
 137             'id': video_id,
 138             'title': title,
 139             'formats': formats,
 140             'thumbnails': thumbnails,
 141             'description': video_data.get('description'),
 142             'duration': float_or_none(video_data.get('duration_ms'), 1000),
 143             'uploader_id': video_data.get('channel_id'),
 144             'series': video_data.get('series_title'),
 145             'season': video_data.get('season_title'),
 146             'season_number': int_or_none(video_data.get('season_number')),
 147             'season_id': video_data.get('season_id'),
 148             'episode': title,
 149             'episode_number': int_or_none(video_data.get('episode_number')),
 150             'episode_id': video_data.get('production_episode_id'),
 151         }