yt_dlp/extractor/discovery.py

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