yt_dlp/extractor/tbs.py

   1 import re
   2 import urllib.parse
   3
   4 from .turner import TurnerBaseIE
   5 from ..utils import (
   6     float_or_none,
   7     int_or_none,
   8     strip_or_none,
   9 )
  10
  11
  12 class TBSIE(TurnerBaseIE):
  13     _VALID_URL = r'https?://(?:www\.)?(?P<site>tbs|tntdrama)\.com(?P<path>/(?:movies|watchtnt|watchtbs|shows/[^/]+/(?:clips|season-\d+/episode-\d+))/(?P<id>[^/?#]+))'
  14     _TESTS = [{
  15         'url': 'http://www.tntdrama.com/shows/the-alienist/clips/monster',
  16         'info_dict': {
  17             'id': '8d384cde33b89f3a43ce5329de42903ed5099887',
  18             'ext': 'mp4',
  19             'title': 'Monster',
  20             'description': 'Get a first look at the theatrical trailer for TNT’s highly anticipated new psychological thriller The Alienist, which premieres January 22 on TNT.',
  21             'timestamp': 1508175329,
  22             'upload_date': '20171016',
  23         },
  24         'params': {
  25             # m3u8 download
  26             'skip_download': True,
  27         },
  28     }, {
  29         'url': 'http://www.tbs.com/shows/search-party/season-1/episode-1/explicit-the-mysterious-disappearance-of-the-girl-no-one-knew',
  30         'only_matching': True,
  31     }, {
  32         'url': 'http://www.tntdrama.com/movies/star-wars-a-new-hope',
  33         'only_matching': True,
  34     }]
  35
  36     def _real_extract(self, url):
  37         site, path, display_id = self._match_valid_url(url).groups()
  38         webpage = self._download_webpage(url, display_id)
  39         drupal_settings = self._parse_json(self._search_regex(
  40             r'<script[^>]+?data-drupal-selector="drupal-settings-json"[^>]*?>({.+?})</script>',
  41             webpage, 'drupal setting'), display_id)
  42         is_live = 'watchtnt' in path or 'watchtbs' in path
  43         video_data = next(v for v in drupal_settings['turner_playlist'] if is_live or v.get('url') == path)
  44
  45         media_id = video_data['mediaID']
  46         title = video_data['title']
  47         tokenizer_query = urllib.parse.parse_qs(urllib.parse.urlparse(
  48             drupal_settings['ngtv_token_url']).query)
  49
  50         info = self._extract_ngtv_info(
  51             media_id, tokenizer_query, {
  52                 'url': url,
  53                 'site_name': site[:3].upper(),
  54                 'auth_required': video_data.get('authRequired') == '1' or is_live,
  55                 'is_live': is_live,
  56             })
  57
  58         thumbnails = []
  59         for image_id, image in video_data.get('images', {}).items():
  60             image_url = image.get('url')
  61             if not image_url or image.get('type') != 'video':
  62                 continue
  63             i = {
  64                 'id': image_id,
  65                 'url': image_url,
  66             }
  67             mobj = re.search(r'(\d+)x(\d+)', image_url)
  68             if mobj:
  69                 i.update({
  70                     'width': int(mobj.group(1)),
  71                     'height': int(mobj.group(2)),
  72                 })
  73             thumbnails.append(i)
  74
  75         info.update({
  76             'id': media_id,
  77             'title': title,
  78             'description': strip_or_none(video_data.get('descriptionNoTags') or video_data.get('shortDescriptionNoTags')),
  79             'duration': float_or_none(video_data.get('duration')) or info.get('duration'),
  80             'timestamp': int_or_none(video_data.get('created')),
  81             'season_number': int_or_none(video_data.get('season')),
  82             'episode_number': int_or_none(video_data.get('episode')),
  83             'thumbnails': thumbnails,
  84             'is_live': is_live,
  85         })
  86         return info