yt_dlp/extractor/tver.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_str
   7 from ..utils import (
   8     int_or_none,
   9     remove_start,
  10     smuggle_url,
  11     try_get,
  12 )
  13
  14
  15 class TVerIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?tver\.jp/(?P<path>(?:corner|episode|feature)/(?P<id>f?\d+))'
  17     # videos are only available for 7 days
  18     _TESTS = [{
  19         'url': 'https://tver.jp/corner/f0062178',
  20         'only_matching': True,
  21     }, {
  22         'url': 'https://tver.jp/feature/f0062413',
  23         'only_matching': True,
  24     }, {
  25         'url': 'https://tver.jp/episode/79622438',
  26         'only_matching': True,
  27     }, {
  28         # subtitle = ' '
  29         'url': 'https://tver.jp/corner/f0068870',
  30         'only_matching': True,
  31     }]
  32     _TOKEN = None
  33     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s'
  34
  35     def _real_initialize(self):
  36         self._TOKEN = self._download_json(
  37             'https://tver.jp/api/access_token.php', None)['token']
  38
  39     def _real_extract(self, url):
  40         path, video_id = self._match_valid_url(url).groups()
  41         main = self._download_json(
  42             'https://api.tver.jp/v4/' + path, video_id,
  43             query={'token': self._TOKEN})['main']
  44         p_id = main['publisher_id']
  45         service = remove_start(main['service'], 'ts_')
  46
  47         r_id = main['reference_id']
  48         if service not in ('tx', 'russia2018', 'sebare2018live', 'gorin'):
  49             r_id = 'ref:' + r_id
  50         bc_url = smuggle_url(
  51             self.BRIGHTCOVE_URL_TEMPLATE % (p_id, r_id),
  52             {'geo_countries': ['JP']})
  53
  54         return {
  55             '_type': 'url_transparent',
  56             'description': try_get(main, lambda x: x['note'][0]['text'], compat_str),
  57             'episode_number': int_or_none(try_get(main, lambda x: x['ext']['episode_number'])),
  58             'url': bc_url,
  59             'ie_key': 'BrightcoveNew',
  60         }