yt_dlp/extractor/nhk.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     clean_html,
   7     get_element_by_class,
   8     int_or_none,
   9     join_nonempty,
  10     parse_duration,
  11     remove_end,
  12     traverse_obj,
  13     try_call,
  14     unescapeHTML,
  15     unified_timestamp,
  16     url_or_none,
  17     urljoin,
  18 )
  19
  20
  21 class NhkBaseIE(InfoExtractor):
  22     _API_URL_TEMPLATE = 'https://nwapi.nhk.jp/nhkworld/%sod%slist/v7b/%s/%s/%s/all%s.json'
  23     _BASE_URL_REGEX = r'https?://www3\.nhk\.or\.jp/nhkworld/(?P<lang>[a-z]{2})/'
  24
  25     def _call_api(self, m_id, lang, is_video, is_episode, is_clip):
  26         return self._download_json(
  27             self._API_URL_TEMPLATE % (
  28                 'v' if is_video else 'r',
  29                 'clip' if is_clip else 'esd',
  30                 'episode' if is_episode else 'program',
  31                 m_id, lang, '/all' if is_video else ''),
  32             m_id, query={'apikey': 'EJfK8jdS57GqlupFgAfAAwr573q01y6k'})['data']['episodes'] or []
  33
  34     def _get_api_info(self, refresh=True):
  35         if not refresh:
  36             return self.cache.load('nhk', 'api_info')
  37
  38         self.cache.store('nhk', 'api_info', {})
  39         movie_player_js = self._download_webpage(
  40             'https://movie-a.nhk.or.jp/world/player/js/movie-player.js', None,
  41             note='Downloading stream API information')
  42         api_info = {
  43             'url': self._search_regex(
  44                 r'prod:[^;]+\bapiUrl:\s*[\'"]([^\'"]+)[\'"]', movie_player_js, None, 'stream API url'),
  45             'token': self._search_regex(
  46                 r'prod:[^;]+\btoken:\s*[\'"]([^\'"]+)[\'"]', movie_player_js, None, 'stream API token'),
  47         }
  48         self.cache.store('nhk', 'api_info', api_info)
  49         return api_info
  50
  51     def _extract_stream_info(self, vod_id):
  52         for refresh in (False, True):
  53             api_info = self._get_api_info(refresh)
  54             if not api_info:
  55                 continue
  56
  57             api_url = api_info.pop('url')
  58             meta = traverse_obj(
  59                 self._download_json(
  60                     api_url, vod_id, 'Downloading stream url info', fatal=False, query={
  61                         **api_info,
  62                         'type': 'json',
  63                         'optional_id': vod_id,
  64                         'active_flg': 1,
  65                     }), ('meta', 0))
  66             stream_url = traverse_obj(
  67                 meta, ('movie_url', ('mb_auto', 'auto_sp', 'auto_pc'), {url_or_none}), get_all=False)
  68
  69             if stream_url:
  70                 formats, subtitles = self._extract_m3u8_formats_and_subtitles(stream_url, vod_id)
  71                 return {
  72                     **traverse_obj(meta, {
  73                         'duration': ('duration', {int_or_none}),
  74                         'timestamp': ('publication_date', {unified_timestamp}),
  75                         'release_timestamp': ('insert_date', {unified_timestamp}),
  76                         'modified_timestamp': ('update_date', {unified_timestamp}),
  77                     }),
  78                     'formats': formats,
  79                     'subtitles': subtitles,
  80                 }
  81         raise ExtractorError('Unable to extract stream url')
  82
  83     def _extract_episode_info(self, url, episode=None):
  84         fetch_episode = episode is None
  85         lang, m_type, episode_id = NhkVodIE._match_valid_url(url).group('lang', 'type', 'id')
  86         is_video = m_type != 'audio'
  87
  88         if is_video:
  89             episode_id = episode_id[:4] + '-' + episode_id[4:]
  90
  91         if fetch_episode:
  92             episode = self._call_api(
  93                 episode_id, lang, is_video, True, episode_id[:4] == '9999')[0]
  94
  95         def get_clean_field(key):
  96             return clean_html(episode.get(key + '_clean') or episode.get(key))
  97
  98         title = get_clean_field('sub_title')
  99         series = get_clean_field('title')
 100
 101         thumbnails = []
 102         for s, w, h in [('', 640, 360), ('_l', 1280, 720)]:
 103             img_path = episode.get('image' + s)
 104             if not img_path:
 105                 continue
 106             thumbnails.append({
 107                 'id': f'{h}p',
 108                 'height': h,
 109                 'width': w,
 110                 'url': 'https://www3.nhk.or.jp' + img_path,
 111             })
 112
 113         episode_name = title
 114         if series and title:
 115             title = f'{series} - {title}'
 116         elif series and not title:
 117             title = series
 118             series = None
 119             episode_name = None
 120         else:  # title, no series
 121             episode_name = None
 122
 123         info = {
 124             'id': episode_id + '-' + lang,
 125             'title': title,
 126             'description': get_clean_field('description'),
 127             'thumbnails': thumbnails,
 128             'series': series,
 129             'episode': episode_name,
 130         }
 131
 132         if is_video:
 133             vod_id = episode['vod_id']
 134             info.update({
 135                 **self._extract_stream_info(vod_id),
 136                 'id': vod_id,
 137             })
 138
 139         else:
 140             if fetch_episode:
 141                 # From https://www3.nhk.or.jp/nhkworld/common/player/radio/inline/rod.html
 142                 audio_path = remove_end(episode['audio']['audio'], '.m4a')
 143                 info['formats'] = self._extract_m3u8_formats(
 144                     f'{urljoin("https://vod-stream.nhk.jp", audio_path)}/index.m3u8',
 145                     episode_id, 'm4a', entry_protocol='m3u8_native',
 146                     m3u8_id='hls', fatal=False)
 147                 for f in info['formats']:
 148                     f['language'] = lang
 149             else:
 150                 info.update({
 151                     '_type': 'url_transparent',
 152                     'ie_key': NhkVodIE.ie_key(),
 153                     'url': url,
 154                 })
 155         return info
 156
 157
 158 class NhkVodIE(NhkBaseIE):
 159     _VALID_URL = [
 160         rf'{NhkBaseIE._BASE_URL_REGEX}shows/(?:(?P<type>video)/)?(?P<id>\d{{4}}[\da-z]\d+)/?(?:$|[?#])',
 161         rf'{NhkBaseIE._BASE_URL_REGEX}(?:ondemand|shows)/(?P<type>audio)/(?P<id>[^/?#]+?-\d{{8}}-[\da-z]+)',
 162         rf'{NhkBaseIE._BASE_URL_REGEX}ondemand/(?P<type>video)/(?P<id>\d{{4}}[\da-z]\d+)',  # deprecated
 163     ]
 164     # Content available only for a limited period of time. Visit
 165     # https://www3.nhk.or.jp/nhkworld/en/ondemand/ for working samples.
 166     _TESTS = [{
 167         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/2049126/',
 168         'info_dict': {
 169             'id': 'nw_vod_v_en_2049_126_20230413233000_01_1681398302',
 170             'ext': 'mp4',
 171             'title': 'Japan Railway Journal - The Tohoku Shinkansen: Full Speed Ahead',
 172             'description': 'md5:49f7c5b206e03868a2fdf0d0814b92f6',
 173             'thumbnail': r're:https://.+/.+\.jpg',
 174             'episode': 'The Tohoku Shinkansen: Full Speed Ahead',
 175             'series': 'Japan Railway Journal',
 176             'modified_timestamp': 1707217907,
 177             'timestamp': 1681428600,
 178             'release_timestamp': 1693883728,
 179             'duration': 1679,
 180             'upload_date': '20230413',
 181             'modified_date': '20240206',
 182             'release_date': '20230905',
 183         },
 184     }, {
 185         # video clip
 186         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/9999011/',
 187         'md5': '153c3016dfd252ba09726588149cf0e7',
 188         'info_dict': {
 189             'id': 'lpZXIwaDE6_Z-976CPsFdxyICyWUzlT5',
 190             'ext': 'mp4',
 191             'title': 'Dining with the Chef - Chef Saito\'s Family recipe: MENCHI-KATSU',
 192             'description': 'md5:5aee4a9f9d81c26281862382103b0ea5',
 193             'thumbnail': r're:https://.+/.+\.jpg',
 194             'series': 'Dining with the Chef',
 195             'episode': 'Chef Saito\'s Family recipe: MENCHI-KATSU',
 196             'duration': 148,
 197             'upload_date': '20190816',
 198             'release_date': '20230902',
 199             'release_timestamp': 1693619292,
 200             'modified_timestamp': 1707217907,
 201             'modified_date': '20240206',
 202             'timestamp': 1565997540,
 203         },
 204     }, {
 205         # radio
 206         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/livinginjapan-20231001-1/',
 207         'info_dict': {
 208             'id': 'livinginjapan-20231001-1-en',
 209             'ext': 'm4a',
 210             'title': 'Living in Japan - Tips for Travelers to Japan / Ramen Vending Machines',
 211             'series': 'Living in Japan',
 212             'description': 'md5:0a0e2077d8f07a03071e990a6f51bfab',
 213             'thumbnail': r're:https://.+/.+\.jpg',
 214             'episode': 'Tips for Travelers to Japan / Ramen Vending Machines',
 215         },
 216     }, {
 217         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/2015173/',
 218         'only_matching': True,
 219     }, {
 220         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/plugin-20190404-1/',
 221         'only_matching': True,
 222     }, {
 223         'url': 'https://www3.nhk.or.jp/nhkworld/fr/ondemand/audio/plugin-20190404-1/',
 224         'only_matching': True,
 225     }, {
 226         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/j_art-20150903-1/',
 227         'only_matching': True,
 228     }, {
 229         # video, alphabetic character in ID #29670
 230         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/9999a34/',
 231         'info_dict': {
 232             'id': 'qfjay6cg',
 233             'ext': 'mp4',
 234             'title': 'DESIGN TALKS plus - Fishermen’s Finery',
 235             'description': 'md5:8a8f958aaafb0d7cb59d38de53f1e448',
 236             'thumbnail': r're:^https?:/(/[a-z0-9.-]+)+\.jpg\?w=1920&h=1080$',
 237             'upload_date': '20210615',
 238             'timestamp': 1623722008,
 239         },
 240         'skip': '404 Not Found',
 241     }, {
 242         # japanese-language, longer id than english
 243         'url': 'https://www3.nhk.or.jp/nhkworld/ja/ondemand/video/0020271111/',
 244         'info_dict': {
 245             'id': 'nw_ja_v_jvod_ohayou_20231008',
 246             'ext': 'mp4',
 247             'title': 'おはよう日本（7時台） - 10月8日放送',
 248             'series': 'おはよう日本（7時台）',
 249             'episode': '10月8日放送',
 250             'thumbnail': r're:https://.+/.+\.jpg',
 251             'description': 'md5:9c1d6cbeadb827b955b20e99ab920ff0',
 252         },
 253         'skip': 'expires 2023-10-15',
 254     }, {
 255         # a one-off (single-episode series). title from the api is just '<p></p>'
 256         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/3004952/',
 257         'info_dict': {
 258             'id': 'nw_vod_v_en_3004_952_20230723091000_01_1690074552',
 259             'ext': 'mp4',
 260             'title': 'Barakan Discovers - AMAMI OSHIMA: Isson\'s Treasure Isla',
 261             'description': 'md5:5db620c46a0698451cc59add8816b797',
 262             'thumbnail': r're:https://.+/.+\.jpg',
 263             'release_date': '20230905',
 264             'timestamp': 1690103400,
 265             'duration': 2939,
 266             'release_timestamp': 1693898699,
 267             'upload_date': '20230723',
 268             'modified_timestamp': 1707217907,
 269             'modified_date': '20240206',
 270             'episode': 'AMAMI OSHIMA: Isson\'s Treasure Isla',
 271             'series': 'Barakan Discovers',
 272         },
 273     }, {
 274         # /ondemand/video/ url with alphabetical character in 5th position of id
 275         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/9999a07/',
 276         'info_dict': {
 277             'id': 'nw_c_en_9999-a07',
 278             'ext': 'mp4',
 279             'episode': 'Mini-Dramas on SDGs: Ep 1 Close the Gender Gap [Director\'s Cut]',
 280             'series': 'Mini-Dramas on SDGs',
 281             'modified_date': '20240206',
 282             'title': 'Mini-Dramas on SDGs - Mini-Dramas on SDGs: Ep 1 Close the Gender Gap [Director\'s Cut]',
 283             'description': 'md5:3f9dcb4db22fceb675d90448a040d3f6',
 284             'timestamp': 1621962360,
 285             'duration': 189,
 286             'release_date': '20230903',
 287             'modified_timestamp': 1707217907,
 288             'upload_date': '20210525',
 289             'thumbnail': r're:https://.+/.+\.jpg',
 290             'release_timestamp': 1693713487,
 291         },
 292     }, {
 293         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/9999d17/',
 294         'info_dict': {
 295             'id': 'nw_c_en_9999-d17',
 296             'ext': 'mp4',
 297             'title': 'Flowers of snow blossom - The 72 Pentads of Yamato',
 298             'description': 'Today’s focus: Snow',
 299             'release_timestamp': 1693792402,
 300             'release_date': '20230904',
 301             'upload_date': '20220128',
 302             'timestamp': 1643370960,
 303             'thumbnail': r're:https://.+/.+\.jpg',
 304             'duration': 136,
 305             'series': '',
 306             'modified_date': '20240206',
 307             'modified_timestamp': 1707217907,
 308         },
 309     }, {
 310         # new /shows/ url format
 311         'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/2032307/',
 312         'info_dict': {
 313             'id': 'nw_vod_v_en_2032_307_20240321113000_01_1710990282',
 314             'ext': 'mp4',
 315             'title': 'Japanology Plus - 20th Anniversary Special Part 1',
 316             'description': 'md5:817d41fc8e54339ad2a916161ea24faf',
 317             'episode': '20th Anniversary Special Part 1',
 318             'series': 'Japanology Plus',
 319             'thumbnail': r're:https://.+/.+\.jpg',
 320             'duration': 1680,
 321             'timestamp': 1711020600,
 322             'upload_date': '20240321',
 323             'release_timestamp': 1711022683,
 324             'release_date': '20240321',
 325             'modified_timestamp': 1711031012,
 326             'modified_date': '20240321',
 327         },
 328     }, {
 329         'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/3020025/',
 330         'info_dict': {
 331             'id': 'nw_vod_v_en_3020_025_20230325144000_01_1679723944',
 332             'ext': 'mp4',
 333             'title': '100 Ideas to Save the World - Working Styles Evolve',
 334             'description': 'md5:9e6c7778eaaf4f7b4af83569649f84d9',
 335             'episode': 'Working Styles Evolve',
 336             'series': '100 Ideas to Save the World',
 337             'thumbnail': r're:https://.+/.+\.jpg',
 338             'duration': 899,
 339             'upload_date': '20230325',
 340             'timestamp': 1679755200,
 341             'release_date': '20230905',
 342             'release_timestamp': 1693880540,
 343             'modified_date': '20240206',
 344             'modified_timestamp': 1707217907,
 345         },
 346     }, {
 347         # new /shows/audio/ url format
 348         'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/audio/livinginjapan-20231001-1/',
 349         'only_matching': True,
 350     }, {
 351         # valid url even if can't be found in wild; support needed for clip entries extraction
 352         'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/9999o80/',
 353         'only_matching': True,
 354     }]
 355
 356     def _real_extract(self, url):
 357         return self._extract_episode_info(url)
 358
 359
 360 class NhkVodProgramIE(NhkBaseIE):
 361     _VALID_URL = rf'''(?x)
 362         {NhkBaseIE._BASE_URL_REGEX}(?:shows|tv)/
 363         (?:(?P<type>audio)/programs/)?(?P<id>\w+)/?
 364         (?:\?(?:[^#]+&)?type=(?P<episode_type>clip|(?:radio|tv)Episode))?'''
 365     _TESTS = [{
 366         # video program episodes
 367         'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/sumo/',
 368         'info_dict': {
 369             'id': 'sumo',
 370             'title': 'GRAND SUMO Highlights',
 371             'description': 'md5:fc20d02dc6ce85e4b72e0273aa52fdbf',
 372         },
 373         'playlist_mincount': 1,
 374     }, {
 375         'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/japanrailway/',
 376         'info_dict': {
 377             'id': 'japanrailway',
 378             'title': 'Japan Railway Journal',
 379             'description': 'md5:ea39d93af7d05835baadf10d1aae0e3f',
 380         },
 381         'playlist_mincount': 12,
 382     }, {
 383         # video program clips
 384         'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/japanrailway/?type=clip',
 385         'info_dict': {
 386             'id': 'japanrailway',
 387             'title': 'Japan Railway Journal',
 388             'description': 'md5:ea39d93af7d05835baadf10d1aae0e3f',
 389         },
 390         'playlist_mincount': 12,
 391     }, {
 392         # audio program
 393         'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/audio/programs/livinginjapan/',
 394         'info_dict': {
 395             'id': 'livinginjapan',
 396             'title': 'Living in Japan',
 397             'description': 'md5:665bb36ec2a12c5a7f598ee713fc2b54',
 398         },
 399         'playlist_mincount': 12,
 400     }, {
 401         # /tv/ program url
 402         'url': 'https://www3.nhk.or.jp/nhkworld/en/tv/designtalksplus/',
 403         'info_dict': {
 404             'id': 'designtalksplus',
 405             'title': 'DESIGN TALKS plus',
 406             'description': 'md5:47b3b3a9f10d4ac7b33b53b70a7d2837',
 407         },
 408         'playlist_mincount': 20,
 409     }, {
 410         'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/10yearshayaomiyazaki/',
 411         'only_matching': True,
 412     }]
 413
 414     @classmethod
 415     def suitable(cls, url):
 416         return False if NhkVodIE.suitable(url) else super().suitable(url)
 417
 418     def _extract_meta_from_class_elements(self, class_values, html):
 419         for class_value in class_values:
 420             if value := clean_html(get_element_by_class(class_value, html)):
 421                 return value
 422
 423     def _real_extract(self, url):
 424         lang, m_type, program_id, episode_type = self._match_valid_url(url).group('lang', 'type', 'id', 'episode_type')
 425         episodes = self._call_api(
 426             program_id, lang, m_type != 'audio', False, episode_type == 'clip')
 427
 428         def entries():
 429             for episode in episodes:
 430                 if episode_path := episode.get('url'):
 431                     yield self._extract_episode_info(urljoin(url, episode_path), episode)
 432
 433         html = self._download_webpage(url, program_id)
 434         program_title = self._extract_meta_from_class_elements([
 435             'p-programDetail__title',  # /ondemand/program/
 436             'pProgramHero__logoText',  # /shows/
 437             'tAudioProgramMain__title',  # /shows/audio/programs/
 438             'p-program-name'], html)  # /tv/
 439         program_description = self._extract_meta_from_class_elements([
 440             'p-programDetail__text',  # /ondemand/program/
 441             'pProgramHero__description',  # /shows/
 442             'tAudioProgramMain__info',  # /shows/audio/programs/
 443             'p-program-description'], html)  # /tv/
 444
 445         return self.playlist_result(entries(), program_id, program_title, program_description)
 446
 447
 448 class NhkForSchoolBangumiIE(InfoExtractor):
 449     _VALID_URL = r'https?://www2\.nhk\.or\.jp/school/movie/(?P<type>bangumi|clip)\.cgi\?das_id=(?P<id>[a-zA-Z0-9_-]+)'
 450     _TESTS = [{
 451         'url': 'https://www2.nhk.or.jp/school/movie/bangumi.cgi?das_id=D0005150191_00000',
 452         'info_dict': {
 453             'id': 'D0005150191_00003',
 454             'title': 'にている かな',
 455             'duration': 599.999,
 456             'timestamp': 1396414800,
 457
 458             'upload_date': '20140402',
 459             'ext': 'mp4',
 460
 461             'chapters': 'count:12',
 462         },
 463         'params': {
 464             # m3u8 download
 465             'skip_download': True,
 466         },
 467     }]
 468
 469     def _real_extract(self, url):
 470         program_type, video_id = self._match_valid_url(url).groups()
 471
 472         webpage = self._download_webpage(
 473             f'https://www2.nhk.or.jp/school/movie/{program_type}.cgi?das_id={video_id}', video_id)
 474
 475         # searches all variables
 476         base_values = {g.group(1): g.group(2) for g in re.finditer(r'var\s+([a-zA-Z_]+)\s*=\s*"([^"]+?)";', webpage)}
 477         # and programObj values too
 478         program_values = {g.group(1): g.group(3) for g in re.finditer(r'(?:program|clip)Obj\.([a-zA-Z_]+)\s*=\s*(["\'])([^"]+?)\2;', webpage)}
 479         # extract all chapters
 480         chapter_durations = [parse_duration(g.group(1)) for g in re.finditer(r'chapterTime\.push\(\'([0-9:]+?)\'\);', webpage)]
 481         chapter_titles = [' '.join([g.group(1) or '', unescapeHTML(g.group(2))]).strip() for g in re.finditer(r'<div class="cpTitle"><span>(scene\s*\d+)?</span>([^<]+?)</div>', webpage)]
 482
 483         # this is how player_core.js is actually doing (!)
 484         version = base_values.get('r_version') or program_values.get('version')
 485         if version:
 486             video_id = f'{video_id.split("_")[0]}_{version}'
 487
 488         formats = self._extract_m3u8_formats(
 489             f'https://nhks-vh.akamaihd.net/i/das/{video_id[0:8]}/{video_id}_V_000.f4v/master.m3u8',
 490             video_id, ext='mp4', m3u8_id='hls')
 491
 492         duration = parse_duration(base_values.get('r_duration'))
 493
 494         chapters = None
 495         if chapter_durations and chapter_titles and len(chapter_durations) == len(chapter_titles):
 496             start_time = chapter_durations
 497             end_time = chapter_durations[1:] + [duration]
 498             chapters = [{
 499                 'start_time': s,
 500                 'end_time': e,
 501                 'title': t,
 502             } for s, e, t in zip(start_time, end_time, chapter_titles)]
 503
 504         return {
 505             'id': video_id,
 506             'title': program_values.get('name'),
 507             'duration': parse_duration(base_values.get('r_duration')),
 508             'timestamp': unified_timestamp(base_values['r_upload']),
 509             'formats': formats,
 510             'chapters': chapters,
 511         }
 512
 513
 514 class NhkForSchoolSubjectIE(InfoExtractor):
 515     IE_DESC = 'Portal page for each school subjects, like Japanese (kokugo, 国語) or math (sansuu/suugaku or 算数・数学)'
 516     KNOWN_SUBJECTS = (
 517         'rika', 'syakai', 'kokugo',
 518         'sansuu', 'seikatsu', 'doutoku',
 519         'ongaku', 'taiiku', 'zukou',
 520         'gijutsu', 'katei', 'sougou',
 521         'eigo', 'tokkatsu',
 522         'tokushi', 'sonota',
 523     )
 524     _VALID_URL = r'https?://www\.nhk\.or\.jp/school/(?P<id>{})/?(?:[\?#].*)?$'.format(
 525         '|'.join(re.escape(s) for s in KNOWN_SUBJECTS))
 526
 527     _TESTS = [{
 528         'url': 'https://www.nhk.or.jp/school/sougou/',
 529         'info_dict': {
 530             'id': 'sougou',
 531             'title': '総合的な学習の時間',
 532         },
 533         'playlist_mincount': 16,
 534     }, {
 535         'url': 'https://www.nhk.or.jp/school/rika/',
 536         'info_dict': {
 537             'id': 'rika',
 538             'title': '理科',
 539         },
 540         'playlist_mincount': 15,
 541     }]
 542
 543     def _real_extract(self, url):
 544         subject_id = self._match_id(url)
 545         webpage = self._download_webpage(url, subject_id)
 546
 547         return self.playlist_from_matches(
 548             re.finditer(rf'href="((?:https?://www\.nhk\.or\.jp)?/school/{re.escape(subject_id)}/[^/]+/)"', webpage),
 549             subject_id,
 550             self._html_search_regex(r'(?s)<span\s+class="subjectName">\s*<img\s*[^<]+>\s*([^<]+?)</span>', webpage, 'title', fatal=False),
 551             lambda g: urljoin(url, g.group(1)))
 552
 553
 554 class NhkForSchoolProgramListIE(InfoExtractor):
 555     _VALID_URL = r'https?://www\.nhk\.or\.jp/school/(?P<id>(?:{})/[a-zA-Z0-9_-]+)'.format(
 556         '|'.join(re.escape(s) for s in NhkForSchoolSubjectIE.KNOWN_SUBJECTS))
 557     _TESTS = [{
 558         'url': 'https://www.nhk.or.jp/school/sougou/q/',
 559         'info_dict': {
 560             'id': 'sougou/q',
 561             'title': 'Ｑ～こどものための哲学',
 562         },
 563         'playlist_mincount': 20,
 564     }]
 565
 566     def _real_extract(self, url):
 567         program_id = self._match_id(url)
 568
 569         webpage = self._download_webpage(f'https://www.nhk.or.jp/school/{program_id}/', program_id)
 570
 571         title = (self._generic_title('', webpage)
 572                  or self._html_search_regex(r'<h3>([^<]+?)とは？\s*</h3>', webpage, 'title', fatal=False))
 573         title = re.sub(r'\s*\|\s*NHK\s+for\s+School\s*$', '', title) if title else None
 574         description = self._html_search_regex(
 575             r'(?s)<div\s+class="programDetail\s*">\s*<p>[^<]+</p>',
 576             webpage, 'description', fatal=False, group=0)
 577
 578         bangumi_list = self._download_json(
 579             f'https://www.nhk.or.jp/school/{program_id}/meta/program.json', program_id)
 580         # they're always bangumi
 581         bangumis = [
 582             self.url_result(f'https://www2.nhk.or.jp/school/movie/bangumi.cgi?das_id={x}')
 583             for x in traverse_obj(bangumi_list, ('part', ..., 'part-video-dasid')) or []]
 584
 585         return self.playlist_result(bangumis, program_id, title, description)
 586
 587
 588 class NhkRadiruIE(InfoExtractor):
 589     _GEO_COUNTRIES = ['JP']
 590     IE_DESC = 'NHK らじる (Radiru/Rajiru)'
 591     _VALID_URL = r'https?://www\.nhk\.or\.jp/radio/(?:player/ondemand|ondemand/detail)\.html\?p=(?P<site>[\da-zA-Z]+)_(?P<corner>[\da-zA-Z]+)(?:_(?P<headline>[\da-zA-Z]+))?'
 592     _TESTS = [{
 593         'url': 'https://www.nhk.or.jp/radio/player/ondemand.html?p=0449_01_3926210',
 594         'skip': 'Episode expired on 2024-02-24',
 595         'info_dict': {
 596             'title': 'ジャズ・トゥナイト　シリーズＪＡＺＺジャイアンツ　５６　ジョニー・ホッジス',
 597             'id': '0449_01_3926210',
 598             'ext': 'm4a',
 599             'series': 'ジャズ・トゥナイト',
 600             'uploader': 'NHK-FM',
 601             'channel': 'NHK-FM',
 602             'thumbnail': 'https://www.nhk.or.jp/prog/img/449/g449.jpg',
 603             'release_date': '20240217',
 604             'description': 'md5:a456ee8e5e59e6dd2a7d32e62386e811',
 605             'timestamp': 1708185600,
 606             'release_timestamp': 1708178400,
 607             'upload_date': '20240217',
 608         },
 609     }, {
 610         # playlist, airs every weekday so it should _hopefully_ be okay forever
 611         'url': 'https://www.nhk.or.jp/radio/ondemand/detail.html?p=0458_01',
 612         'info_dict': {
 613             'id': '0458_01',
 614             'title': 'ベストオブクラシック',
 615             'description': '世界中の上質な演奏会をじっくり堪能する本格派クラシック番組。',
 616             'channel': 'NHK-FM',
 617             'uploader': 'NHK-FM',
 618             'thumbnail': 'https://www.nhk.or.jp/prog/img/458/g458.jpg',
 619         },
 620         'playlist_mincount': 3,
 621     }, {
 622         # one with letters in the id
 623         'url': 'https://www.nhk.or.jp/radio/player/ondemand.html?p=F300_06_3738470',
 624         'note': 'Expires on 2024-03-31',
 625         'info_dict': {
 626             'id': 'F300_06_3738470',
 627             'ext': 'm4a',
 628             'title': '有島武郎「一房のぶどう」',
 629             'description': '朗読：川野一宇（ラジオ深夜便アンカー）\r\n\r\n（2016年12月8日放送「ラジオ深夜便『アンカー朗読シリーズ』」より）',
 630             'channel': 'NHKラジオ第1、NHK-FM',
 631             'uploader': 'NHKラジオ第1、NHK-FM',
 632             'timestamp': 1635757200,
 633             'thumbnail': 'https://www.nhk.or.jp/radioondemand/json/F300/img/corner/box_109_thumbnail.jpg',
 634             'release_date': '20161207',
 635             'series': 'らじる文庫 by ラジオ深夜便 ',
 636             'release_timestamp': 1481126700,
 637             'upload_date': '20211101',
 638         },
 639         'expected_warnings': ['Unable to download JSON metadata', 'Failed to get extended description'],
 640     }, {
 641         # news
 642         'url': 'https://www.nhk.or.jp/radio/player/ondemand.html?p=F261_01_3855109',
 643         'skip': 'Expires on 2023-04-17',
 644         'info_dict': {
 645             'id': 'F261_01_3855109',
 646             'ext': 'm4a',
 647             'channel': 'NHKラジオ第1',
 648             'uploader': 'NHKラジオ第1',
 649             'timestamp': 1681635900,
 650             'release_date': '20230416',
 651             'series': 'NHKラジオニュース',
 652             'title': '午後６時のNHKニュース',
 653             'thumbnail': 'https://www.nhk.or.jp/radioondemand/json/F261/img/RADIONEWS_640.jpg',
 654             'upload_date': '20230416',
 655             'release_timestamp': 1681635600,
 656         },
 657     }]
 658
 659     _API_URL_TMPL = None
 660
 661     def _extract_extended_description(self, episode_id, episode):
 662         service, _, area = traverse_obj(episode, ('aa_vinfo2', {str}, {lambda x: (x or '').partition(',')}))
 663         aa_vinfo3 = traverse_obj(episode, ('aa_vinfo3', {str}))
 664         detail_url = try_call(
 665             lambda: self._API_URL_TMPL.format(service=service, area=area, dateid=aa_vinfo3))
 666         if not detail_url:
 667             return
 668
 669         full_meta = traverse_obj(
 670             self._download_json(detail_url, episode_id, 'Downloading extended metadata', fatal=False),
 671             ('list', service, 0, {dict})) or {}
 672         return join_nonempty('subtitle', 'content', 'act', 'music', delim='\n\n', from_dict=full_meta)
 673
 674     def _extract_episode_info(self, headline, programme_id, series_meta):
 675         episode_id = f'{programme_id}_{headline["headline_id"]}'
 676         episode = traverse_obj(headline, ('file_list', 0, {dict}))
 677         description = self._extract_extended_description(episode_id, episode)
 678         if not description:
 679             self.report_warning('Failed to get extended description, falling back to summary')
 680             description = traverse_obj(episode, ('file_title_sub', {str}))
 681
 682         return {
 683             **series_meta,
 684             'id': episode_id,
 685             'formats': self._extract_m3u8_formats(episode.get('file_name'), episode_id, fatal=False),
 686             'container': 'm4a_dash',  # force fixup, AAC-only HLS
 687             'was_live': True,
 688             'series': series_meta.get('title'),
 689             'thumbnail': url_or_none(headline.get('headline_image')) or series_meta.get('thumbnail'),
 690             'description': description,
 691             **traverse_obj(episode, {
 692                 'title': 'file_title',
 693                 'timestamp': ('open_time', {unified_timestamp}),
 694                 'release_timestamp': ('aa_vinfo4', {lambda x: x.split('_')[0]}, {unified_timestamp}),
 695             }),
 696         }
 697
 698     def _real_initialize(self):
 699         if self._API_URL_TMPL:
 700             return
 701         api_config = self._download_xml(
 702             'https://www.nhk.or.jp/radio/config/config_web.xml', None, 'Downloading API config', fatal=False)
 703         NhkRadiruIE._API_URL_TMPL = try_call(lambda: f'https:{api_config.find(".//url_program_detail").text}')
 704
 705     def _real_extract(self, url):
 706         site_id, corner_id, headline_id = self._match_valid_url(url).group('site', 'corner', 'headline')
 707         programme_id = f'{site_id}_{corner_id}'
 708
 709         if site_id == 'F261':
 710             json_url = 'https://www.nhk.or.jp/s-media/news/news-site/list/v1/all.json'
 711         else:
 712             json_url = f'https://www.nhk.or.jp/radioondemand/json/{site_id}/bangumi_{programme_id}.json'
 713
 714         meta = self._download_json(json_url, programme_id)['main']
 715
 716         series_meta = traverse_obj(meta, {
 717             'title': 'program_name',
 718             'channel': 'media_name',
 719             'uploader': 'media_name',
 720             'thumbnail': (('thumbnail_c', 'thumbnail_p'), {url_or_none}),
 721         }, get_all=False)
 722
 723         if headline_id:
 724             return self._extract_episode_info(
 725                 traverse_obj(meta, (
 726                     'detail_list', lambda _, v: v['headline_id'] == headline_id), get_all=False),
 727                 programme_id, series_meta)
 728
 729         def entries():
 730             for headline in traverse_obj(meta, ('detail_list', ..., {dict})):
 731                 yield self._extract_episode_info(headline, programme_id, series_meta)
 732
 733         return self.playlist_result(
 734             entries(), programme_id, playlist_description=meta.get('site_detail'), **series_meta)
 735
 736
 737 class NhkRadioNewsPageIE(InfoExtractor):
 738     _VALID_URL = r'https?://www\.nhk\.or\.jp/radionews/?(?:$|[?#])'
 739     _TESTS = [{
 740         # airs daily, on-the-hour most hours
 741         'url': 'https://www.nhk.or.jp/radionews/',
 742         'playlist_mincount': 5,
 743         'info_dict': {
 744             'id': 'F261_01',
 745             'thumbnail': 'https://www.nhk.or.jp/radioondemand/json/F261/img/RADIONEWS_640.jpg',
 746             'description': 'md5:bf2c5b397e44bc7eb26de98d8f15d79d',
 747             'channel': 'NHKラジオ第1',
 748             'uploader': 'NHKラジオ第1',
 749             'title': 'NHKラジオニュース',
 750         },
 751     }]
 752
 753     def _real_extract(self, url):
 754         return self.url_result('https://www.nhk.or.jp/radio/ondemand/detail.html?p=F261_01', NhkRadiruIE)
 755
 756
 757 class NhkRadiruLiveIE(InfoExtractor):
 758     _GEO_COUNTRIES = ['JP']
 759     _VALID_URL = r'https?://www\.nhk\.or\.jp/radio/player/\?ch=(?P<id>r[12]|fm)'
 760     _TESTS = [{
 761         # radio 1, no area specified
 762         'url': 'https://www.nhk.or.jp/radio/player/?ch=r1',
 763         'info_dict': {
 764             'id': 'r1-tokyo',
 765             'title': 're:^ＮＨＫネットラジオ第1 東京.+$',
 766             'ext': 'm4a',
 767             'thumbnail': 'https://www.nhk.or.jp/common/img/media/r1-200x200.png',
 768             'live_status': 'is_live',
 769         },
 770     }, {
 771         # radio 2, area specified
 772         # (the area doesnt actually matter, r2 is national)
 773         'url': 'https://www.nhk.or.jp/radio/player/?ch=r2',
 774         'params': {'extractor_args': {'nhkradirulive': {'area': ['fukuoka']}}},
 775         'info_dict': {
 776             'id': 'r2-fukuoka',
 777             'title': 're:^ＮＨＫネットラジオ第2 福岡.+$',
 778             'ext': 'm4a',
 779             'thumbnail': 'https://www.nhk.or.jp/common/img/media/r2-200x200.png',
 780             'live_status': 'is_live',
 781         },
 782     }, {
 783         # fm, area specified
 784         'url': 'https://www.nhk.or.jp/radio/player/?ch=fm',
 785         'params': {'extractor_args': {'nhkradirulive': {'area': ['sapporo']}}},
 786         'info_dict': {
 787             'id': 'fm-sapporo',
 788             'title': 're:^ＮＨＫネットラジオＦＭ 札幌.+$',
 789             'ext': 'm4a',
 790             'thumbnail': 'https://www.nhk.or.jp/common/img/media/fm-200x200.png',
 791             'live_status': 'is_live',
 792         },
 793     }]
 794
 795     _NOA_STATION_IDS = {'r1': 'n1', 'r2': 'n2', 'fm': 'n3'}
 796
 797     def _real_extract(self, url):
 798         station = self._match_id(url)
 799         area = self._configuration_arg('area', ['tokyo'])[0]
 800
 801         config = self._download_xml(
 802             'https://www.nhk.or.jp/radio/config/config_web.xml', station, 'Downloading area information')
 803         data = config.find(f'.//data//area[.="{area}"]/..')
 804
 805         if not data:
 806             raise ExtractorError('Invalid area. Valid areas are: {}'.format(', '.join(
 807                 [i.text for i in config.findall('.//data//area')])), expected=True)
 808
 809         noa_info = self._download_json(
 810             f'https:{config.find(".//url_program_noa").text}'.format(area=data.find('areakey').text),
 811             station, note=f'Downloading {area} station metadata', fatal=False)
 812         present_info = traverse_obj(noa_info, ('nowonair_list', self._NOA_STATION_IDS.get(station), 'present'))
 813
 814         return {
 815             'title': ' '.join(traverse_obj(present_info, (('service', 'area'), 'name', {str}))),
 816             'id': join_nonempty(station, area),
 817             'thumbnails': traverse_obj(present_info, ('service', 'images', ..., {
 818                 'url': 'url',
 819                 'width': ('width', {int_or_none}),
 820                 'height': ('height', {int_or_none}),
 821             })),
 822             'formats': self._extract_m3u8_formats(data.find(f'{station}hls').text, station),
 823             'is_live': True,
 824         }