yt_dlp/extractor/tver.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     join_nonempty,
   5     smuggle_url,
   6     str_or_none,
   7     strip_or_none,
   8     traverse_obj,
   9 )
  10
  11
  12 class TVerIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?tver\.jp/(?:(?P<type>lp|corner|series|episodes?|feature|tokyo2020/video)/)+(?P<id>[a-zA-Z0-9]+)'
  14     _TESTS = [{
  15         'skip': 'videos are only available for 7 days',
  16         'url': 'https://tver.jp/episodes/ep83nf3w4p',
  17         'info_dict': {
  18             'title': '家事ヤロウ!!! 売り場席巻のチーズSP＆財前直見×森泉親子の脱東京暮らし密着！',
  19             'description': 'md5:dc2c06b6acc23f1e7c730c513737719b',
  20             'series': '家事ヤロウ!!!',
  21             'episode': '売り場席巻のチーズSP＆財前直見×森泉親子の脱東京暮らし密着！',
  22             'alt_title': '売り場席巻のチーズSP＆財前直見×森泉親子の脱東京暮らし密着！',
  23             'channel': 'テレビ朝日',
  24             'onair_label': '5月3日(火)放送分',
  25             'ext_title': '家事ヤロウ!!! 売り場席巻のチーズSP＆財前直見×森泉親子の脱東京暮らし密着！ テレビ朝日 5月3日(火)放送分',
  26         },
  27         'add_ie': ['BrightcoveNew'],
  28     }, {
  29         'url': 'https://tver.jp/corner/f0103888',
  30         'only_matching': True,
  31     }, {
  32         'url': 'https://tver.jp/lp/f0033031',
  33         'only_matching': True,
  34     }]
  35     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s'
  36     _PLATFORM_UID = None
  37     _PLATFORM_TOKEN = None
  38
  39     def _real_initialize(self):
  40         create_response = self._download_json(
  41             'https://platform-api.tver.jp/v2/api/platform_users/browser/create', None,
  42             note='Creating session', data=b'device_type=pc', headers={
  43                 'Origin': 'https://s.tver.jp',
  44                 'Referer': 'https://s.tver.jp/',
  45                 'Content-Type': 'application/x-www-form-urlencoded',
  46             })
  47         self._PLATFORM_UID = traverse_obj(create_response, ('result', 'platform_uid'))
  48         self._PLATFORM_TOKEN = traverse_obj(create_response, ('result', 'platform_token'))
  49
  50     def _real_extract(self, url):
  51         video_id, video_type = self._match_valid_url(url).group('id', 'type')
  52         if video_type not in {'series', 'episodes'}:
  53             webpage = self._download_webpage(url, video_id, note='Resolving to new URL')
  54             video_id = self._match_id(self._search_regex(
  55                 (r'canonical"\s*href="(https?://tver\.jp/[^"]+)"', r'&link=(https?://tver\.jp/[^?&]+)[?&]'),
  56                 webpage, 'url regex'))
  57         video_info = self._download_json(
  58             f'https://statics.tver.jp/content/episode/{video_id}.json', video_id,
  59             query={'v': '5'}, headers={
  60                 'Origin': 'https://tver.jp',
  61                 'Referer': 'https://tver.jp/',
  62             })
  63         p_id = video_info['video']['accountID']
  64         r_id = traverse_obj(video_info, ('video', ('videoRefID', 'videoID')), get_all=False)
  65         if not r_id:
  66             raise ExtractorError('Failed to extract reference ID for Brightcove')
  67         if not r_id.isdigit():
  68             r_id = f'ref:{r_id}'
  69
  70         additional_info = self._download_json(
  71             f'https://platform-api.tver.jp/service/api/v1/callEpisode/{video_id}?require_data=mylist,later[epefy106ur],good[epefy106ur],resume[epefy106ur]',
  72             video_id, fatal=False,
  73             query={
  74                 'platform_uid': self._PLATFORM_UID,
  75                 'platform_token': self._PLATFORM_TOKEN,
  76             }, headers={
  77                 'x-tver-platform-type': 'web'
  78             })
  79
  80         additional_content_info = traverse_obj(
  81             additional_info, ('result', 'episode', 'content'), get_all=False) or {}
  82         episode = strip_or_none(additional_content_info.get('title'))
  83         series = str_or_none(additional_content_info.get('seriesTitle'))
  84         title = (
  85             join_nonempty(series, episode, delim=' ')
  86             or str_or_none(video_info.get('title')))
  87         provider = str_or_none(additional_content_info.get('productionProviderName'))
  88         onair_label = str_or_none(additional_content_info.get('broadcastDateLabel'))
  89
  90         return {
  91             '_type': 'url_transparent',
  92             'title': title,
  93             'series': series,
  94             'episode': episode,
  95             # an another title which is considered "full title" for some viewers
  96             'alt_title': join_nonempty(title, provider, onair_label, delim=' '),
  97             'channel': provider,
  98             'description': str_or_none(video_info.get('description')),
  99             'url': smuggle_url(
 100                 self.BRIGHTCOVE_URL_TEMPLATE % (p_id, r_id), {'geo_countries': ['JP']}),
 101             'ie_key': 'BrightcoveNew',
 102         }