yt_dlp/extractor/tbsjp.py

   1 from .common import InfoExtractor
   2 from ..networking.exceptions import HTTPError
   3 from ..utils import (
   4     ExtractorError,
   5     clean_html,
   6     get_element_text_and_html_by_tag,
   7     int_or_none,
   8     str_or_none,
   9     traverse_obj,
  10     try_call,
  11     unified_timestamp,
  12     urljoin,
  13 )
  14
  15
  16 class TBSJPEpisodeIE(InfoExtractor):
  17     _VALID_URL = r'https?://cu\.tbs\.co\.jp/episode/(?P<id>[\d_]+)'
  18     _GEO_BYPASS = False
  19     _TESTS = [{
  20         'url': 'https://cu.tbs.co.jp/episode/23613_2044134_1000049010',
  21         'skip': 'streams geo-restricted, Japan only. Also, will likely expire eventually',
  22         'info_dict': {
  23             'title': 'VIVANT 第三話 誤送金完結へ!絶体絶命の反撃開始',
  24             'id': '23613_2044134_1000049010',
  25             'ext': 'mp4',
  26             'upload_date': '20230728',
  27             'duration': 3517,
  28             'release_timestamp': 1691118230,
  29             'episode': '第三話 誤送金完結へ!絶体絶命の反撃開始',
  30             'release_date': '20230804',
  31             'categories': 'count:11',
  32             'episode_number': 3,
  33             'timestamp': 1690522538,
  34             'description': 'md5:2b796341af1ef772034133174ba4a895',
  35             'series': 'VIVANT',
  36         },
  37     }]
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41         webpage = self._download_webpage(url, video_id)
  42         meta = self._search_json(r'window\.app\s*=', webpage, 'episode info', video_id, fatal=False)
  43         episode = traverse_obj(meta, ('falcorCache', 'catalog', 'episode', video_id, 'value'))
  44
  45         tf_path = self._search_regex(
  46             r'<script[^>]+src=["\'](/assets/tf\.[^"\']+\.js)["\']', webpage, 'stream API config')
  47         tf_js = self._download_webpage(urljoin(url, tf_path), video_id, note='Downloading stream API config')
  48         video_url = self._search_regex(r'videoPlaybackUrl:\s*[\'"]([^\'"]+)[\'"]', tf_js, 'stream API url')
  49         api_key = self._search_regex(r'api_key:\s*[\'"]([^\'"]+)[\'"]', tf_js, 'stream API key')
  50
  51         try:
  52             source_meta = self._download_json(f'{video_url}ref:{video_id}', video_id,
  53                                               headers={'X-Streaks-Api-Key': api_key},
  54                                               note='Downloading stream metadata')
  55         except ExtractorError as e:
  56             if isinstance(e.cause, HTTPError) and e.cause.status == 403:
  57                 self.raise_geo_restricted(countries=['JP'])
  58             raise
  59
  60         formats, subtitles = [], {}
  61         for src in traverse_obj(source_meta, ('sources', ..., 'src')):
  62             fmts, subs = self._extract_m3u8_formats_and_subtitles(src, video_id, fatal=False)
  63             formats.extend(fmts)
  64             self._merge_subtitles(subs, target=subtitles)
  65
  66         return {
  67             'title': try_call(lambda: clean_html(get_element_text_and_html_by_tag('h3', webpage)[0])),
  68             'id': video_id,
  69             **traverse_obj(episode, {
  70                 'categories': ('keywords', {list}),
  71                 'id': ('content_id', {str}),
  72                 'description': ('description', 0, 'value'),
  73                 'timestamp': ('created_at', {unified_timestamp}),
  74                 'release_timestamp': ('pub_date', {unified_timestamp}),
  75                 'duration': ('tv_episode_info', 'duration', {int_or_none}),
  76                 'episode_number': ('tv_episode_info', 'episode_number', {int_or_none}),
  77                 'episode': ('title', lambda _, v: not v.get('is_phonetic'), 'value'),
  78                 'series': ('custom_data', 'program_name'),
  79             }, get_all=False),
  80             'formats': formats,
  81             'subtitles': subtitles,
  82         }
  83
  84
  85 class TBSJPProgramIE(InfoExtractor):
  86     _VALID_URL = r'https?://cu\.tbs\.co\.jp/program/(?P<id>\d+)'
  87     _TESTS = [{
  88         'url': 'https://cu.tbs.co.jp/program/23601',
  89         'playlist_mincount': 4,
  90         'info_dict': {
  91             'id': '23601',
  92             'categories': ['エンタメ', 'ミライカプセル', '会社', '働く', 'バラエティ', '動画'],
  93             'description': '幼少期の夢は大人になって、どう成長したのだろうか？\nそしてその夢は今後、どのように広がっていくのか？\nいま話題の会社で働く人の「夢の成長」を描く',
  94             'series': 'ミライカプセル　-I have a dream-',
  95             'title': 'ミライカプセル　-I have a dream-',
  96         },
  97     }]
  98
  99     def _real_extract(self, url):
 100         programme_id = self._match_id(url)
 101         webpage = self._download_webpage(url, programme_id)
 102         meta = self._search_json(r'window\.app\s*=', webpage, 'programme info', programme_id)
 103
 104         programme = traverse_obj(meta, ('falcorCache', 'catalog', 'program', programme_id, 'false', 'value'))
 105
 106         return {
 107             '_type': 'playlist',
 108             'entries': [self.url_result(f'https://cu.tbs.co.jp/episode/{video_id}', TBSJPEpisodeIE, video_id)
 109                         for video_id in traverse_obj(programme, ('custom_data', 'seriesList', 'episodeCode', ...))],
 110             'id': programme_id,
 111             **traverse_obj(programme, {
 112                 'categories': ('keywords', ...),
 113                 'id': ('tv_episode_info', 'show_content_id', {str_or_none}),
 114                 'description': ('custom_data', 'program_description'),
 115                 'series': ('custom_data', 'program_name'),
 116                 'title': ('custom_data', 'program_name'),
 117             }),
 118         }
 119
 120
 121 class TBSJPPlaylistIE(InfoExtractor):
 122     _VALID_URL = r'https?://cu\.tbs\.co\.jp/playlist/(?P<id>[\da-f]+)'
 123     _TESTS = [{
 124         'url': 'https://cu.tbs.co.jp/playlist/184f9970e7ba48e4915f1b252c55015e',
 125         'playlist_mincount': 4,
 126         'info_dict': {
 127             'title': 'まもなく配信終了',
 128             'id': '184f9970e7ba48e4915f1b252c55015e',
 129         },
 130     }]
 131
 132     def _real_extract(self, url):
 133         playlist_id = self._match_id(url)
 134         page = self._download_webpage(url, playlist_id)
 135         meta = self._search_json(r'window\.app\s*=', page, 'playlist info', playlist_id)
 136         playlist = traverse_obj(meta, ('falcorCache', 'playList', playlist_id))
 137
 138         def entries():
 139             for entry in traverse_obj(playlist, ('catalogs', 'value', lambda _, v: v['content_id'])):
 140                 # TODO: it's likely possible to get all metadata from the playlist page json instead
 141                 content_id = entry['content_id']
 142                 content_type = entry.get('content_type')
 143                 if content_type == 'tv_show':
 144                     yield self.url_result(
 145                         f'https://cu.tbs.co.jp/program/{content_id}', TBSJPProgramIE, content_id)
 146                 elif content_type == 'tv_episode':
 147                     yield self.url_result(
 148                         f'https://cu.tbs.co.jp/episode/{content_id}', TBSJPEpisodeIE, content_id)
 149                 else:
 150                     self.report_warning(f'Skipping "{content_id}" with unsupported content_type "{content_type}"')
 151
 152         return self.playlist_result(entries(), playlist_id, traverse_obj(playlist, ('display_name', 'value')))