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