yt_dlp/extractor/discovery.py

   1 from __future__ import unicode_literals
   2
   3 import random
   4 import string
   5
   6 from .discoverygo import DiscoveryGoBaseIE
   7 from ..compat import compat_urllib_parse_unquote
   8 from ..utils import ExtractorError
   9 from ..compat import compat_HTTPError
  10
  11
  12 class DiscoveryIE(DiscoveryGoBaseIE):
  13     _VALID_URL = r'''(?x)https?://
  14         (?P<site>
  15             go\.discovery|
  16             www\.
  17                 (?:
  18                     investigationdiscovery|
  19                     discoverylife|
  20                     animalplanet|
  21                     ahctv|
  22                     destinationamerica|
  23                     sciencechannel|
  24                     tlc
  25                 )|
  26             watch\.
  27                 (?:
  28                     hgtv|
  29                     foodnetwork|
  30                     travelchannel|
  31                     diynetwork|
  32                     cookingchanneltv|
  33                     motortrend
  34                 )
  35         )\.com/tv-shows/(?P<show_slug>[^/]+)/(?:video|full-episode)s/(?P<id>[^./?#]+)'''
  36     _TESTS = [{
  37         'url': 'https://go.discovery.com/tv-shows/cash-cab/videos/riding-with-matthew-perry',
  38         'info_dict': {
  39             'id': '5a2f35ce6b66d17a5026e29e',
  40             'ext': 'mp4',
  41             'title': 'Riding with Matthew Perry',
  42             'description': 'md5:a34333153e79bc4526019a5129e7f878',
  43             'duration': 84,
  44         },
  45         'params': {
  46             'skip_download': True,  # requires ffmpeg
  47         }
  48     }, {
  49         'url': 'https://www.investigationdiscovery.com/tv-shows/final-vision/full-episodes/final-vision',
  50         'only_matching': True,
  51     }, {
  52         'url': 'https://go.discovery.com/tv-shows/alaskan-bush-people/videos/follow-your-own-road',
  53         'only_matching': True,
  54     }, {
  55         # using `show_slug` is important to get the correct video data
  56         'url': 'https://www.sciencechannel.com/tv-shows/mythbusters-on-science/full-episodes/christmas-special',
  57         'only_matching': True,
  58     }]
  59     _GEO_COUNTRIES = ['US']
  60     _GEO_BYPASS = False
  61     _API_BASE_URL = 'https://api.discovery.com/v1/'
  62
  63     def _real_extract(self, url):
  64         site, show_slug, display_id = self._match_valid_url(url).groups()
  65
  66         access_token = None
  67         cookies = self._get_cookies(url)
  68
  69         # prefer Affiliate Auth Token over Anonymous Auth Token
  70         auth_storage_cookie = cookies.get('eosAf') or cookies.get('eosAn')
  71         if auth_storage_cookie and auth_storage_cookie.value:
  72             auth_storage = self._parse_json(compat_urllib_parse_unquote(
  73                 compat_urllib_parse_unquote(auth_storage_cookie.value)),
  74                 display_id, fatal=False) or {}
  75             access_token = auth_storage.get('a') or auth_storage.get('access_token')
  76
  77         if not access_token:
  78             access_token = self._download_json(
  79                 'https://%s.com/anonymous' % site, display_id,
  80                 'Downloading token JSON metadata', query={
  81                     'authRel': 'authorization',
  82                     'client_id': '3020a40c2356a645b4b4',
  83                     'nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
  84                     'redirectUri': 'https://www.discovery.com/',
  85                 })['access_token']
  86
  87         headers = self.geo_verification_headers()
  88         headers['Authorization'] = 'Bearer ' + access_token
  89
  90         try:
  91             video = self._download_json(
  92                 self._API_BASE_URL + 'content/videos',
  93                 display_id, 'Downloading content JSON metadata',
  94                 headers=headers, query={
  95                     'embed': 'show.name',
  96                     'fields': 'authenticated,description.detailed,duration,episodeNumber,id,name,parental.rating,season.number,show,tags',
  97                     'slug': display_id,
  98                     'show_slug': show_slug,
  99                 })[0]
 100             video_id = video['id']
 101             stream = self._download_json(
 102                 self._API_BASE_URL + 'streaming/video/' + video_id,
 103                 display_id, 'Downloading streaming JSON metadata', headers=headers)
 104         except ExtractorError as e:
 105             if isinstance(e.cause, compat_HTTPError) and e.cause.code in (401, 403):
 106                 e_description = self._parse_json(
 107                     e.cause.read().decode(), display_id)['description']
 108                 if 'resource not available for country' in e_description:
 109                     self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
 110                 if 'Authorized Networks' in e_description:
 111                     raise ExtractorError(
 112                         'This video is only available via cable service provider subscription that'
 113                         ' is not currently supported. You may want to use --cookies.', expected=True)
 114                 raise ExtractorError(e_description)
 115             raise
 116
 117         return self._extract_video_info(video, stream, display_id)