yt_dlp/extractor/tver.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     int_or_none,
  10     remove_start,
  11     smuggle_url,
  12     try_get,
  13 )
  14
  15
  16 class TVerIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?tver\.jp/(?P<path>(?:corner|episode|feature)/(?P<id>f?\d+))'
  18     # videos are only available for 7 days
  19     _TESTS = [{
  20         'url': 'https://tver.jp/corner/f0062178',
  21         'only_matching': True,
  22     }, {
  23         'url': 'https://tver.jp/feature/f0062413',
  24         'only_matching': True,
  25     }, {
  26         'url': 'https://tver.jp/episode/79622438',
  27         'only_matching': True,
  28     }, {
  29         # subtitle = ' '
  30         'url': 'https://tver.jp/corner/f0068870',
  31         'only_matching': True,
  32     }]
  33     _TOKEN = None
  34     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s'
  35
  36     def _real_initialize(self):
  37         self._TOKEN = self._download_json(
  38             'https://tver.jp/api/access_token.php', None)['token']
  39
  40     def _real_extract(self, url):
  41         path, video_id = re.match(self._VALID_URL, url).groups()
  42         main = self._download_json(
  43             'https://api.tver.jp/v4/' + path, video_id,
  44             query={'token': self._TOKEN})['main']
  45         p_id = main['publisher_id']
  46         service = remove_start(main['service'], 'ts_')
  47
  48         r_id = main['reference_id']
  49         if service not in ('tx', 'russia2018', 'sebare2018live', 'gorin'):
  50             r_id = 'ref:' + r_id
  51         bc_url = smuggle_url(
  52             self.BRIGHTCOVE_URL_TEMPLATE % (p_id, r_id),
  53             {'geo_countries': ['JP']})
  54
  55         return {
  56             '_type': 'url_transparent',
  57             'description': try_get(main, lambda x: x['note'][0]['text'], compat_str),
  58             'episode_number': int_or_none(try_get(main, lambda x: x['ext']['episode_number'])),
  59             'url': bc_url,
  60             'ie_key': 'BrightcoveNew',
  61         }