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 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             (?:(?:www|go)\.)?discovery|
  17             (?:www\.)?
  18                 (?:
  19                     investigationdiscovery|
  20                     discoverylife|
  21                     animalplanet|
  22                     ahctv|
  23                     destinationamerica|
  24                     sciencechannel|
  25                     tlc|
  26                     velocity
  27                 )|
  28             watch\.
  29                 (?:
  30                     hgtv|
  31                     foodnetwork|
  32                     travelchannel|
  33                     diynetwork|
  34                     cookingchanneltv|
  35                     motortrend
  36                 )
  37         )\.com/tv-shows/(?P<show_slug>[^/]+)/(?:video|full-episode)s/(?P<id>[^./?#]+)'''
  38     _TESTS = [{
  39         'url': 'https://go.discovery.com/tv-shows/cash-cab/videos/riding-with-matthew-perry',
  40         'info_dict': {
  41             'id': '5a2f35ce6b66d17a5026e29e',
  42             'ext': 'mp4',
  43             'title': 'Riding with Matthew Perry',
  44             'description': 'md5:a34333153e79bc4526019a5129e7f878',
  45             'duration': 84,
  46         },
  47         'params': {
  48             'skip_download': True,  # requires ffmpeg
  49         }
  50     }, {
  51         'url': 'https://www.investigationdiscovery.com/tv-shows/final-vision/full-episodes/final-vision',
  52         'only_matching': True,
  53     }, {
  54         'url': 'https://go.discovery.com/tv-shows/alaskan-bush-people/videos/follow-your-own-road',
  55         'only_matching': True,
  56     }, {
  57         # using `show_slug` is important to get the correct video data
  58         'url': 'https://www.sciencechannel.com/tv-shows/mythbusters-on-science/full-episodes/christmas-special',
  59         'only_matching': True,
  60     }]
  61     _GEO_COUNTRIES = ['US']
  62     _GEO_BYPASS = False
  63     _API_BASE_URL = 'https://api.discovery.com/v1/'
  64
  65     def _real_extract(self, url):
  66         site, show_slug, display_id = re.match(self._VALID_URL, url).groups()
  67
  68         access_token = None
  69         cookies = self._get_cookies(url)
  70
  71         # prefer Affiliate Auth Token over Anonymous Auth Token
  72         auth_storage_cookie = cookies.get('eosAf') or cookies.get('eosAn')
  73         if auth_storage_cookie and auth_storage_cookie.value:
  74             auth_storage = self._parse_json(compat_urllib_parse_unquote(
  75                 compat_urllib_parse_unquote(auth_storage_cookie.value)),
  76                 display_id, fatal=False) or {}
  77             access_token = auth_storage.get('a') or auth_storage.get('access_token')
  78
  79         if not access_token:
  80             access_token = self._download_json(
  81                 'https://%s.com/anonymous' % site, display_id,
  82                 'Downloading token JSON metadata', query={
  83                     'authRel': 'authorization',
  84                     'client_id': '3020a40c2356a645b4b4',
  85                     'nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
  86                     'redirectUri': 'https://fusion.ddmcdn.com/app/mercury-sdk/180/redirectHandler.html?https://www.%s.com' % site,
  87                 })['access_token']
  88
  89         headers = self.geo_verification_headers()
  90         headers['Authorization'] = 'Bearer ' + access_token
  91
  92         try:
  93             video = self._download_json(
  94                 self._API_BASE_URL + 'content/videos',
  95                 display_id, 'Downloading content JSON metadata',
  96                 headers=headers, query={
  97                     'embed': 'show.name',
  98                     'fields': 'authenticated,description.detailed,duration,episodeNumber,id,name,parental.rating,season.number,show,tags',
  99                     'slug': display_id,
 100                     'show_slug': show_slug,
 101                 })[0]
 102             video_id = video['id']
 103             stream = self._download_json(
 104                 self._API_BASE_URL + 'streaming/video/' + video_id,
 105                 display_id, 'Downloading streaming JSON metadata', headers=headers)
 106         except ExtractorError as e:
 107             if isinstance(e.cause, compat_HTTPError) and e.cause.code in (401, 403):
 108                 e_description = self._parse_json(
 109                     e.cause.read().decode(), display_id)['description']
 110                 if 'resource not available for country' in e_description:
 111                     self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
 112                 if 'Authorized Networks' in e_description:
 113                     raise ExtractorError(
 114                         'This video is only available via cable service provider subscription that'
 115                         ' is not currently supported. You may want to use --cookies.', expected=True)
 116                 raise ExtractorError(e_description)
 117             raise
 118
 119         return self._extract_video_info(video, stream, display_id)