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     ExtractorError,
   9     int_or_none,
  10     remove_start,
  11     smuggle_url,
  12     traverse_obj,
  13 )
  14
  15
  16 class TVerIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?tver\.jp/(?P<path>corner|episode|feature|lp|tokyo2020/video)/(?P<id>[fc]?\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         'url': 'https://tver.jp/lp/f0009694',
  34         'only_matching': True,
  35     }, {
  36         'url': 'https://tver.jp/lp/c0000239',
  37         'only_matching': True,
  38     }, {
  39         'url': 'https://tver.jp/tokyo2020/video/6264525510001',
  40         'only_matching': True,
  41     }]
  42     _TOKEN = None
  43     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s'
  44
  45     def _real_initialize(self):
  46         self._TOKEN = self._download_json(
  47             'https://tver.jp/api/access_token.php', None)['token']
  48
  49     def _real_extract(self, url):
  50         path, video_id = self._match_valid_url(url).groups()
  51         if path == 'lp':
  52             webpage = self._download_webpage(url, video_id)
  53             redirect_path = self._search_regex(r'to_href="([^"]+)', webpage, 'redirect path')
  54             path, video_id = self._match_valid_url(f'https://tver.jp{redirect_path}').groups()
  55         api_response = self._download_json(f'https://api.tver.jp/v4/{path}/{video_id}', video_id, query={'token': self._TOKEN})
  56         p_id = traverse_obj(api_response, ('main', 'publisher_id'))
  57         if not p_id:
  58             error_msg, expected = traverse_obj(api_response, ('episode', 0, 'textbar', 0, ('text', 'longer')), get_all=False), True
  59             if not error_msg:
  60                 error_msg, expected = 'Failed to extract publisher ID', False
  61             raise ExtractorError(error_msg, expected=expected)
  62         service = remove_start(traverse_obj(api_response, ('main', 'service')), 'ts_')
  63
  64         r_id = traverse_obj(api_response, ('main', 'reference_id'))
  65         if service not in ('tx', 'russia2018', 'sebare2018live', 'gorin'):
  66             r_id = 'ref:' + r_id
  67         bc_url = smuggle_url(
  68             self.BRIGHTCOVE_URL_TEMPLATE % (p_id, r_id),
  69             {'geo_countries': ['JP']})
  70
  71         return {
  72             '_type': 'url_transparent',
  73             'description': traverse_obj(api_response, ('main', 'note', 0, 'text'), expected_type=compat_str),
  74             'episode_number': int_or_none(traverse_obj(api_response, ('main', 'ext', 'episode_number'), expected_type=compat_str)),
  75             'url': bc_url,
  76             'ie_key': 'BrightcoveNew',
  77         }