Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/discovery.py

   1 from __future__ import unicode_literals
   2
   3 import random
   4 import re
   5 import string
   6
   7 from .discoverygo import DiscoveryGoBaseIE
   8 from ..compat import (
   9     compat_str,
  10     compat_urllib_parse_unquote,
  11 )
  12 from ..utils import (
  13     ExtractorError,
  14     try_get,
  15 )
  16 from ..compat import compat_HTTPError
  17
  18
  19 class DiscoveryIE(DiscoveryGoBaseIE):
  20     _VALID_URL = r'''(?x)https?://(?:www\.)?(?P<site>
  21             discovery|
  22             investigationdiscovery|
  23             discoverylife|
  24             animalplanet|
  25             ahctv|
  26             destinationamerica|
  27             sciencechannel|
  28             tlc|
  29             velocity
  30         )\.com(?P<path>/tv-shows/[^/]+/(?:video|full-episode)s/(?P<id>[^./?#]+))'''
  31     _TESTS = [{
  32         'url': 'https://www.discovery.com/tv-shows/cash-cab/videos/dave-foley',
  33         'info_dict': {
  34             'id': '5a2d9b4d6b66d17a5026e1fd',
  35             'ext': 'mp4',
  36             'title': 'Dave Foley',
  37             'description': 'md5:4b39bcafccf9167ca42810eb5f28b01f',
  38             'duration': 608,
  39         },
  40         'params': {
  41             'skip_download': True,  # requires ffmpeg
  42         }
  43     }, {
  44         'url': 'https://www.investigationdiscovery.com/tv-shows/final-vision/full-episodes/final-vision',
  45         'only_matching': True,
  46     }]
  47     _GEO_COUNTRIES = ['US']
  48     _GEO_BYPASS = False
  49
  50     def _real_extract(self, url):
  51         site, path, display_id = re.match(self._VALID_URL, url).groups()
  52         webpage = self._download_webpage(url, display_id)
  53
  54         react_data = self._parse_json(self._search_regex(
  55             r'window\.__reactTransmitPacket\s*=\s*({.+?});',
  56             webpage, 'react data'), display_id)
  57         content_blocks = react_data['layout'][path]['contentBlocks']
  58         video = next(cb for cb in content_blocks if cb.get('type') == 'video')['content']['items'][0]
  59         video_id = video['id']
  60
  61         access_token = None
  62         cookies = self._get_cookies(url)
  63
  64         # prefer Affiliate Auth Token over Anonymous Auth Token
  65         auth_storage_cookie = cookies.get('eosAf') or cookies.get('eosAn')
  66         if auth_storage_cookie and auth_storage_cookie.value:
  67             auth_storage = self._parse_json(compat_urllib_parse_unquote(
  68                 compat_urllib_parse_unquote(auth_storage_cookie.value)),
  69                 video_id, fatal=False) or {}
  70             access_token = auth_storage.get('a') or auth_storage.get('access_token')
  71
  72         if not access_token:
  73             access_token = self._download_json(
  74                 'https://www.%s.com/anonymous' % site, display_id, query={
  75                     'authRel': 'authorization',
  76                     'client_id': try_get(
  77                         react_data, lambda x: x['application']['apiClientId'],
  78                         compat_str) or '3020a40c2356a645b4b4',
  79                     'nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
  80                     'redirectUri': 'https://fusion.ddmcdn.com/app/mercury-sdk/180/redirectHandler.html?https://www.%s.com' % site,
  81                 })['access_token']
  82
  83         try:
  84             stream = self._download_json(
  85                 'https://api.discovery.com/v1/streaming/video/' + video_id,
  86                 display_id, headers={
  87                     'Authorization': 'Bearer ' + access_token,
  88                 })
  89         except ExtractorError as e:
  90             if isinstance(e.cause, compat_HTTPError) and e.cause.code in (401, 403):
  91                 e_description = self._parse_json(
  92                     e.cause.read().decode(), display_id)['description']
  93                 if 'resource not available for country' in e_description:
  94                     self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
  95                 if 'Authorized Networks' in e_description:
  96                     raise ExtractorError(
  97                         'This video is only available via cable service provider subscription that'
  98                         ' is not currently supported. You may want to use --cookies.', expected=True)
  99                 raise ExtractorError(e_description)
 100             raise
 101
 102         return self._extract_video_info(video, stream, display_id)