yt_dlp/extractor/stv.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     float_or_none,
   4     int_or_none,
   5     smuggle_url,
   6     str_or_none,
   7     try_get,
   8 )
   9
  10
  11 class STVPlayerIE(InfoExtractor):
  12     IE_NAME = 'stv:player'
  13     _VALID_URL = r'https?://player\.stv\.tv/(?P<type>episode|video)/(?P<id>[a-z0-9]{4})'
  14     _TESTS = [{
  15         # shortform
  16         'url': 'https://player.stv.tv/video/4gwd/emmerdale/60-seconds-on-set-with-laura-norton/',
  17         'md5': '5adf9439c31d554f8be0707c7abe7e0a',
  18         'info_dict': {
  19             'id': '5333973339001',
  20             'ext': 'mp4',
  21             'upload_date': '20170301',
  22             'title': '60 seconds on set with Laura Norton',
  23             'description': "How many questions can Laura - a.k.a Kerry Wyatt - answer in 60 seconds? Let's find out!",
  24             'timestamp': 1488388054,
  25             'uploader_id': '1486976045',
  26         },
  27         'skip': 'this resource is unavailable outside of the UK',
  28     }, {
  29         # episodes
  30         'url': 'https://player.stv.tv/episode/4125/jennifer-saunders-memory-lane',
  31         'only_matching': True,
  32     }]
  33     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1486976045/default_default/index.html?videoId=%s'
  34     _PTYPE_MAP = {
  35         'episode': 'episodes',
  36         'video': 'shortform',
  37     }
  38
  39     def _real_extract(self, url):
  40         ptype, video_id = self._match_valid_url(url).groups()
  41
  42         webpage = self._download_webpage(url, video_id, fatal=False) or ''
  43         props = self._search_nextjs_data(webpage, video_id, default={}).get('props') or {}
  44         player_api_cache = try_get(
  45             props, lambda x: x['initialReduxState']['playerApiCache']) or {}
  46
  47         api_path, resp = None, {}
  48         for k, v in player_api_cache.items():
  49             if k.startswith(('/episodes/', '/shortform/')):
  50                 api_path, resp = k, v
  51                 break
  52         else:
  53             episode_id = str_or_none(try_get(
  54                 props, lambda x: x['pageProps']['episodeId']))
  55             api_path = f'/{self._PTYPE_MAP[ptype]}/{episode_id or video_id}'
  56
  57         result = resp.get('results')
  58         if not result:
  59             resp = self._download_json(
  60                 'https://player.api.stv.tv/v1' + api_path, video_id)
  61             result = resp['results']
  62
  63         video = result['video']
  64         video_id = str(video['id'])
  65
  66         subtitles = {}
  67         _subtitles = result.get('_subtitles') or {}
  68         for ext, sub_url in _subtitles.items():
  69             subtitles.setdefault('en', []).append({
  70                 'ext': 'vtt' if ext == 'webvtt' else ext,
  71                 'url': sub_url,
  72             })
  73
  74         programme = result.get('programme') or {}
  75         if programme.get('drmEnabled'):
  76             self.report_drm(video_id)
  77
  78         return {
  79             '_type': 'url_transparent',
  80             'id': video_id,
  81             'url': smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % video_id, {'geo_countries': ['GB']}),
  82             'description': result.get('summary'),
  83             'duration': float_or_none(video.get('length'), 1000),
  84             'subtitles': subtitles,
  85             'view_count': int_or_none(result.get('views')),
  86             'series': programme.get('name') or programme.get('shortName'),
  87             'ie_key': 'BrightcoveNew',
  88         }