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_str
   9 from ..utils import (
  10     ExtractorError,
  11     try_get,
  12 )
  13 from ..compat import compat_HTTPError
  14
  15
  16 class DiscoveryIE(DiscoveryGoBaseIE):
  17     _VALID_URL = r'''(?x)https?://(?:www\.)?(?P<site>
  18             discovery|
  19             investigationdiscovery|
  20             discoverylife|
  21             animalplanet|
  22             ahctv|
  23             destinationamerica|
  24             sciencechannel|
  25             tlc|
  26             velocity
  27         )\.com(?P<path>/tv-shows/[^/]+/(?:video|full-episode)s/(?P<id>[^./?#]+))'''
  28     _TESTS = [{
  29         'url': 'https://www.discovery.com/tv-shows/cash-cab/videos/dave-foley',
  30         'info_dict': {
  31             'id': '5a2d9b4d6b66d17a5026e1fd',
  32             'ext': 'mp4',
  33             'title': 'Dave Foley',
  34             'description': 'md5:4b39bcafccf9167ca42810eb5f28b01f',
  35             'duration': 608,
  36         },
  37         'params': {
  38             'skip_download': True,  # requires ffmpeg
  39         }
  40     }, {
  41         'url': 'https://www.investigationdiscovery.com/tv-shows/final-vision/full-episodes/final-vision',
  42         'only_matching': True,
  43     }]
  44     _GEO_COUNTRIES = ['US']
  45     _GEO_BYPASS = False
  46
  47     def _real_extract(self, url):
  48         site, path, display_id = re.match(self._VALID_URL, url).groups()
  49         webpage = self._download_webpage(url, display_id)
  50
  51         react_data = self._parse_json(self._search_regex(
  52             r'window\.__reactTransmitPacket\s*=\s*({.+?});',
  53             webpage, 'react data'), display_id)
  54         content_blocks = react_data['layout'][path]['contentBlocks']
  55         video = next(cb for cb in content_blocks if cb.get('type') == 'video')['content']['items'][0]
  56         video_id = video['id']
  57
  58         access_token = self._download_json(
  59             'https://www.%s.com/anonymous' % site, display_id, query={
  60                 'authRel': 'authorization',
  61                 'client_id': try_get(
  62                     react_data, lambda x: x['application']['apiClientId'],
  63                     compat_str) or '3020a40c2356a645b4b4',
  64                 'nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
  65                 'redirectUri': 'https://fusion.ddmcdn.com/app/mercury-sdk/180/redirectHandler.html?https://www.%s.com' % site,
  66             })['access_token']
  67
  68         try:
  69             stream = self._download_json(
  70                 'https://api.discovery.com/v1/streaming/video/' + video_id,
  71                 display_id, headers={
  72                     'Authorization': 'Bearer ' + access_token,
  73                 })
  74         except ExtractorError as e:
  75             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
  76                 e_description = self._parse_json(
  77                     e.cause.read().decode(), display_id)['description']
  78                 if 'resource not available for country' in e_description:
  79                     self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
  80                 if 'Authorized Networks' in e_description:
  81                     raise ExtractorError(
  82                         'This video is only available via cable service provider subscription that'
  83                         ' is not currently supported. You may want to use --cookies.', expected=True)
  84                 raise ExtractorError(e_description)
  85             raise
  86
  87         return self._extract_video_info(video, stream, display_id)