[yt-dlp.git] / yt_dlp / extractor / nhk.py

from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..utils import (
    parse_duration,
    traverse_obj,
    unescapeHTML,
    unified_timestamp,
    urljoin
)


class NhkBaseIE(InfoExtractor):
    _API_URL_TEMPLATE = 'https://api.nhk.or.jp/nhkworld/%sod%slist/v7a/%s/%s/%s/all%s.json'
    _BASE_URL_REGEX = r'https?://www3\.nhk\.or\.jp/nhkworld/(?P<lang>[a-z]{2})/ondemand'
    _TYPE_REGEX = r'/(?P<type>video|audio)/'

    def _call_api(self, m_id, lang, is_video, is_episode, is_clip):
        return self._download_json(
            self._API_URL_TEMPLATE % (
                'v' if is_video else 'r',
                'clip' if is_clip else 'esd',
                'episode' if is_episode else 'program',
                m_id, lang, '/all' if is_video else ''),
            m_id, query={'apikey': 'EJfK8jdS57GqlupFgAfAAwr573q01y6k'})['data']['episodes'] or []

    def _extract_episode_info(self, url, episode=None):
        fetch_episode = episode is None
        lang, m_type, episode_id = NhkVodIE._match_valid_url(url).groups()
        if episode_id.isdigit():
            episode_id = episode_id[:4] + '-' + episode_id[4:]

        is_video = m_type == 'video'
        if fetch_episode:
            episode = self._call_api(
                episode_id, lang, is_video, True, episode_id[:4] == '9999')[0]
        title = episode.get('sub_title_clean') or episode['sub_title']

        def get_clean_field(key):
            return episode.get(key + '_clean') or episode.get(key)

        series = get_clean_field('title')

        thumbnails = []
        for s, w, h in [('', 640, 360), ('_l', 1280, 720)]:
            img_path = episode.get('image' + s)
            if not img_path:
                continue
            thumbnails.append({
                'id': '%dp' % h,
                'height': h,
                'width': w,
                'url': 'https://www3.nhk.or.jp' + img_path,
            })

        info = {
            'id': episode_id + '-' + lang,
            'title': '%s - %s' % (series, title) if series and title else title,
            'description': get_clean_field('description'),
            'thumbnails': thumbnails,
            'series': series,
            'episode': title,
        }
        if is_video:
            vod_id = episode['vod_id']
            info.update({
                '_type': 'url_transparent',
                'ie_key': 'Piksel',
                'url': 'https://player.piksel.com/v/refid/nhkworld/prefid/' + vod_id,
                'id': vod_id,
            })
        else:
            if fetch_episode:
                audio_path = episode['audio']['audio']
                info['formats'] = self._extract_m3u8_formats(
                    'https://nhkworld-vh.akamaihd.net/i%s/master.m3u8' % audio_path,
                    episode_id, 'm4a', entry_protocol='m3u8_native',
                    m3u8_id='hls', fatal=False)
                for f in info['formats']:
                    f['language'] = lang
                self._sort_formats(info['formats'])
            else:
                info.update({
                    '_type': 'url_transparent',
                    'ie_key': NhkVodIE.ie_key(),
                    'url': url,
                })
        return info


class NhkVodIE(NhkBaseIE):
    _VALID_URL = r'%s%s(?P<id>\d{7}|[^/]+?-\d{8}-[0-9a-z]+)' % (NhkBaseIE._BASE_URL_REGEX, NhkBaseIE._TYPE_REGEX)
    # Content available only for a limited period of time. Visit
    # https://www3.nhk.or.jp/nhkworld/en/ondemand/ for working samples.
    _TESTS = [{
        # video clip
        'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/9999011/',
        'md5': '7a90abcfe610ec22a6bfe15bd46b30ca',
        'info_dict': {
            'id': 'a95j5iza',
            'ext': 'mp4',
            'title': "Dining with the Chef - Chef Saito's Family recipe: MENCHI-KATSU",
            'description': 'md5:5aee4a9f9d81c26281862382103b0ea5',
            'timestamp': 1565965194,
            'upload_date': '20190816',
        },
    }, {
        # audio clip
        'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/r_inventions-20201104-1/',
        'info_dict': {
            'id': 'r_inventions-20201104-1-en',
            'ext': 'm4a',
            'title': "Japan's Top Inventions - Miniature Video Cameras",
            'description': 'md5:07ea722bdbbb4936fdd360b6a480c25b',
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }, {
        'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/2015173/',
        'only_matching': True,
    }, {
        'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/plugin-20190404-1/',
        'only_matching': True,
    }, {
        'url': 'https://www3.nhk.or.jp/nhkworld/fr/ondemand/audio/plugin-20190404-1/',
        'only_matching': True,
    }, {
        'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/j_art-20150903-1/',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        return self._extract_episode_info(url)


class NhkVodProgramIE(NhkBaseIE):
    _VALID_URL = r'%s/program%s(?P<id>[0-9a-z]+)(?:.+?\btype=(?P<episode_type>clip|(?:radio|tv)Episode))?' % (NhkBaseIE._BASE_URL_REGEX, NhkBaseIE._TYPE_REGEX)
    _TESTS = [{
        # video program episodes
        'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/japanrailway',
        'info_dict': {
            'id': 'japanrailway',
            'title': 'Japan Railway Journal',
        },
        'playlist_mincount': 1,
    }, {
        # video program clips
        'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/japanrailway/?type=clip',
        'info_dict': {
            'id': 'japanrailway',
            'title': 'Japan Railway Journal',
        },
        'playlist_mincount': 5,
    }, {
        'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/10yearshayaomiyazaki/',
        'only_matching': True,
    }, {
        # audio program
        'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/audio/listener/',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        lang, m_type, program_id, episode_type = self._match_valid_url(url).groups()

        episodes = self._call_api(
            program_id, lang, m_type == 'video', False, episode_type == 'clip')

        entries = []
        for episode in episodes:
            episode_path = episode.get('url')
            if not episode_path:
                continue
            entries.append(self._extract_episode_info(
                urljoin(url, episode_path), episode))

        program_title = None
        if entries:
            program_title = entries[0].get('series')

        return self.playlist_result(entries, program_id, program_title)


class NhkForSchoolBangumiIE(InfoExtractor):
    _VALID_URL = r'https?://www2\.nhk\.or\.jp/school/movie/(?P<type>bangumi|clip)\.cgi\?das_id=(?P<id>[a-zA-Z0-9_-]+)'
    _TESTS = [{
        'url': 'https://www2.nhk.or.jp/school/movie/bangumi.cgi?das_id=D0005150191_00000',
        'info_dict': {
            'id': 'D0005150191_00003',
            'title': 'にている かな',
            'duration': 599.999,
            'timestamp': 1396414800,

            'upload_date': '20140402',
            'ext': 'mp4',

            'chapters': 'count:12'
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }]

    def _real_extract(self, url):
        program_type, video_id = self._match_valid_url(url).groups()

        webpage = self._download_webpage(
            f'https://www2.nhk.or.jp/school/movie/{program_type}.cgi?das_id={video_id}', video_id)

        # searches all variables
        base_values = {g.group(1): g.group(2) for g in re.finditer(r'var\s+([a-zA-Z_]+)\s*=\s*"([^"]+?)";', webpage)}
        # and programObj values too
        program_values = {g.group(1): g.group(3) for g in re.finditer(r'(?:program|clip)Obj\.([a-zA-Z_]+)\s*=\s*(["\'])([^"]+?)\2;', webpage)}
        # extract all chapters
        chapter_durations = [parse_duration(g.group(1)) for g in re.finditer(r'chapterTime\.push\(\'([0-9:]+?)\'\);', webpage)]
        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)]

        # this is how player_core.js is actually doing (!)
        version = base_values.get('r_version') or program_values.get('version')
        if version:
            video_id = f'{video_id.split("_")[0]}_{version}'

        formats = self._extract_m3u8_formats(
            f'https://nhks-vh.akamaihd.net/i/das/{video_id[0:8]}/{video_id}_V_000.f4v/master.m3u8',
            video_id, ext='mp4', m3u8_id='hls')
        self._sort_formats(formats)

        duration = parse_duration(base_values.get('r_duration'))

        chapters = None
        if chapter_durations and chapter_titles and len(chapter_durations) == len(chapter_titles):
            start_time = chapter_durations
            end_time = chapter_durations[1:] + [duration]
            chapters = [{
                'start_time': s,
                'end_time': e,
                'title': t,
            } for s, e, t in zip(start_time, end_time, chapter_titles)]

        return {
            'id': video_id,
            'title': program_values.get('name'),
            'duration': parse_duration(base_values.get('r_duration')),
            'timestamp': unified_timestamp(base_values['r_upload']),
            'formats': formats,
            'chapters': chapters,
        }


class NhkForSchoolSubjectIE(InfoExtractor):
    IE_DESC = 'Portal page for each school subjects, like Japanese (kokugo, 国語) or math (sansuu/suugaku or 算数・数学)'
    KNOWN_SUBJECTS = (
        'rika', 'syakai', 'kokugo',
        'sansuu', 'seikatsu', 'doutoku',
        'ongaku', 'taiiku', 'zukou',
        'gijutsu', 'katei', 'sougou',
        'eigo', 'tokkatsu',
        'tokushi', 'sonota',
    )
    _VALID_URL = r'https?://www\.nhk\.or\.jp/school/(?P<id>%s)/?(?:[\?#].*)?$' % '|'.join(re.escape(s) for s in KNOWN_SUBJECTS)

    _TESTS = [{
        'url': 'https://www.nhk.or.jp/school/sougou/',
        'info_dict': {
            'id': 'sougou',
            'title': '総合的な学習の時間',
        },
        'playlist_mincount': 16,
    }, {
        'url': 'https://www.nhk.or.jp/school/rika/',
        'info_dict': {
            'id': 'rika',
            'title': '理科',
        },
        'playlist_mincount': 15,
    }]

    def _real_extract(self, url):
        subject_id = self._match_id(url)
        webpage = self._download_webpage(url, subject_id)

        return self.playlist_from_matches(
            re.finditer(rf'href="((?:https?://www\.nhk\.or\.jp)?/school/{re.escape(subject_id)}/[^/]+/)"', webpage),
            subject_id,
            self._html_search_regex(r'(?s)<span\s+class="subjectName">\s*<img\s*[^<]+>\s*([^<]+?)</span>', webpage, 'title', fatal=False),
            lambda g: urljoin(url, g.group(1)))


class NhkForSchoolProgramListIE(InfoExtractor):
    _VALID_URL = r'https?://www\.nhk\.or\.jp/school/(?P<id>(?:%s)/[a-zA-Z0-9_-]+)' % (
        '|'.join(re.escape(s) for s in NhkForSchoolSubjectIE.KNOWN_SUBJECTS)
    )
    _TESTS = [{
        'url': 'https://www.nhk.or.jp/school/sougou/q/',
        'info_dict': {
            'id': 'sougou/q',
            'title': 'Ｑ～こどものための哲学',
        },
        'playlist_mincount': 20,
    }]

    def _real_extract(self, url):
        program_id = self._match_id(url)

        webpage = self._download_webpage(f'https://www.nhk.or.jp/school/{program_id}/', program_id)

        title = (self._og_search_title(webpage)
                 or self._html_extract_title(webpage)
                 or self._html_search_regex(r'<h3>([^<]+?)とは？\s*</h3>', webpage, 'title', fatal=False))
        title = re.sub(r'\s*\|\s*NHK\s+for\s+School\s*$', '', title) if title else None
        description = self._html_search_regex(
            r'(?s)<div\s+class="programDetail\s*">\s*<p>[^<]+</p>',
            webpage, 'description', fatal=False, group=0)

        bangumi_list = self._download_json(
            f'https://www.nhk.or.jp/school/{program_id}/meta/program.json', program_id)
        # they're always bangumi
        bangumis = [
            self.url_result(f'https://www2.nhk.or.jp/school/movie/bangumi.cgi?das_id={x}')
            for x in traverse_obj(bangumi_list, ('part', ..., 'part-video-dasid')) or []]

        return self.playlist_result(bangumis, program_id, title, description)
Commit	Line	Data
298a120a AN	1	from __future__ import unicode_literals
298a120a AN	2
77cc7c6e	3	import re
061d1cd9	4
298a120a	5	from .common import InfoExtractor
77cc7c6e LNO	6	from ..utils import (
	7	parse_duration,
	8	traverse_obj,
	9	unescapeHTML,
	10	unified_timestamp,
	11	urljoin
	12	)
298a120a AN	13
298a120a AN	14
29f7c58a	15	class NhkBaseIE(InfoExtractor):
	16	_API_URL_TEMPLATE = 'https://api.nhk.or.jp/nhkworld/%sod%slist/v7a/%s/%s/%s/all%s.json'
	17	_BASE_URL_REGEX = r'https?://www3\.nhk\.or\.jp/nhkworld/(?P<lang>[a-z]{2})/ondemand'
	18	_TYPE_REGEX = r'/(?P<type>video\|audio)/'
298a120a	19
29f7c58a	20	def _call_api(self, m_id, lang, is_video, is_episode, is_clip):
	21	return self._download_json(
	22	self._API_URL_TEMPLATE % (
	23	'v' if is_video else 'r',
	24	'clip' if is_clip else 'esd',
	25	'episode' if is_episode else 'program',
	26	m_id, lang, '/all' if is_video else ''),
	27	m_id, query={'apikey': 'EJfK8jdS57GqlupFgAfAAwr573q01y6k'})['data']['episodes'] or []
	28
	29	def _extract_episode_info(self, url, episode=None):
	30	fetch_episode = episode is None
5ad28e7f	31	lang, m_type, episode_id = NhkVodIE._match_valid_url(url).groups()
061d1cd9 RA	32	if episode_id.isdigit():
061d1cd9 RA	33	episode_id = episode_id[:4] + '-' + episode_id[4:]
f9b373af	34
061d1cd9	35	is_video = m_type == 'video'
29f7c58a	36	if fetch_episode:
	37	episode = self._call_api(
	38	episode_id, lang, is_video, True, episode_id[:4] == '9999')[0]
061d1cd9	39	title = episode.get('sub_title_clean') or episode['sub_title']
45396dd2	40
061d1cd9 RA	41	def get_clean_field(key):
061d1cd9 RA	42	return episode.get(key + '_clean') or episode.get(key)
45396dd2	43
061d1cd9	44	series = get_clean_field('title')
45396dd2	45
061d1cd9 RA	46	thumbnails = []
	47	for s, w, h in [('', 640, 360), ('_l', 1280, 720)]:
	48	img_path = episode.get('image' + s)
	49	if not img_path:
	50	continue
	51	thumbnails.append({
	52	'id': '%dp' % h,
	53	'height': h,
	54	'width': w,
	55	'url': 'https://www3.nhk.or.jp' + img_path,
	56	})
298a120a	57
061d1cd9 RA	58	info = {
061d1cd9 RA	59	'id': episode_id + '-' + lang,
f9b373af	60	'title': '%s - %s' % (series, title) if series and title else title,
061d1cd9 RA	61	'description': get_clean_field('description'),
061d1cd9 RA	62	'thumbnails': thumbnails,
f9b373af S	63	'series': series,
	64	'episode': title,
	65	}
061d1cd9	66	if is_video:
29f7c58a	67	vod_id = episode['vod_id']
061d1cd9 RA	68	info.update({
061d1cd9 RA	69	'_type': 'url_transparent',
a373befa	70	'ie_key': 'Piksel',
29f7c58a	71	'url': 'https://player.piksel.com/v/refid/nhkworld/prefid/' + vod_id,
29f7c58a	72	'id': vod_id,
061d1cd9 RA	73	})
061d1cd9 RA	74	else:
29f7c58a	75	if fetch_episode:
	76	audio_path = episode['audio']['audio']
	77	info['formats'] = self._extract_m3u8_formats(
	78	'https://nhkworld-vh.akamaihd.net/i%s/master.m3u8' % audio_path,
	79	episode_id, 'm4a', entry_protocol='m3u8_native',
	80	m3u8_id='hls', fatal=False)
	81	for f in info['formats']:
	82	f['language'] = lang
df6c409d	83	self._sort_formats(info['formats'])
29f7c58a	84	else:
	85	info.update({
	86	'_type': 'url_transparent',
	87	'ie_key': NhkVodIE.ie_key(),
	88	'url': url,
	89	})
061d1cd9	90	return info
29f7c58a	91
	92
	93	class NhkVodIE(NhkBaseIE):
	94	_VALID_URL = r'%s%s(?P<id>\d{7}\|[^/]+?-\d{8}-[0-9a-z]+)' % (NhkBaseIE._BASE_URL_REGEX, NhkBaseIE._TYPE_REGEX)
	95	# Content available only for a limited period of time. Visit
	96	# https://www3.nhk.or.jp/nhkworld/en/ondemand/ for working samples.
	97	_TESTS = [{
	98	# video clip
	99	'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/9999011/',
	100	'md5': '7a90abcfe610ec22a6bfe15bd46b30ca',
	101	'info_dict': {
	102	'id': 'a95j5iza',
	103	'ext': 'mp4',
	104	'title': "Dining with the Chef - Chef Saito's Family recipe: MENCHI-KATSU",
	105	'description': 'md5:5aee4a9f9d81c26281862382103b0ea5',
	106	'timestamp': 1565965194,
	107	'upload_date': '20190816',
	108	},
	109	}, {
	110	# audio clip
	111	'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/r_inventions-20201104-1/',
	112	'info_dict': {
	113	'id': 'r_inventions-20201104-1-en',
	114	'ext': 'm4a',
	115	'title': "Japan's Top Inventions - Miniature Video Cameras",
	116	'description': 'md5:07ea722bdbbb4936fdd360b6a480c25b',
	117	},
	118	'params': {
	119	# m3u8 download
	120	'skip_download': True,
	121	},
	122	}, {
	123	'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/2015173/',
	124	'only_matching': True,
	125	}, {
	126	'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/plugin-20190404-1/',
	127	'only_matching': True,
	128	}, {
	129	'url': 'https://www3.nhk.or.jp/nhkworld/fr/ondemand/audio/plugin-20190404-1/',
	130	'only_matching': True,
	131	}, {
	132	'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/j_art-20150903-1/',
	133	'only_matching': True,
	134	}]
	135
	136	def _real_extract(self, url):
	137	return self._extract_episode_info(url)
	138
	139
	140	class NhkVodProgramIE(NhkBaseIE):
	141	_VALID_URL = r'%s/program%s(?P<id>[0-9a-z]+)(?:.+?\btype=(?P<episode_type>clip\|(?:radio\|tv)Episode))?' % (NhkBaseIE._BASE_URL_REGEX, NhkBaseIE._TYPE_REGEX)
	142	_TESTS = [{
	143	# video program episodes
	144	'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/japanrailway',
	145	'info_dict': {
	146	'id': 'japanrailway',
	147	'title': 'Japan Railway Journal',
	148	},
	149	'playlist_mincount': 1,
	150	}, {
	151	# video program clips
	152	'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/japanrailway/?type=clip',
	153	'info_dict': {
	154	'id': 'japanrailway',
155	'title': 'Japan Railway Journal',
156	},
157	'playlist_mincount': 5,
158	}, {
159	'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/10yearshayaomiyazaki/',
160	'only_matching': True,
161	}, {
162	# audio program
163	'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/audio/listener/',
164	'only_matching': True,
165	}]
166
167	def _real_extract(self, url):
5ad28e7f	168	lang, m_type, program_id, episode_type = self._match_valid_url(url).groups()
29f7c58a	169
	170	episodes = self._call_api(
	171	program_id, lang, m_type == 'video', False, episode_type == 'clip')
	172
	173	entries = []
	174	for episode in episodes:
	175	episode_path = episode.get('url')
	176	if not episode_path:
	177	continue
	178	entries.append(self._extract_episode_info(
	179	urljoin(url, episode_path), episode))
	180
	181	program_title = None
	182	if entries:
	183	program_title = entries[0].get('series')
	184
	185	return self.playlist_result(entries, program_id, program_title)
77cc7c6e LNO	186
	187
	188	class NhkForSchoolBangumiIE(InfoExtractor):
	189	_VALID_URL = r'https?://www2\.nhk\.or\.jp/school/movie/(?P<type>bangumi\|clip)\.cgi\?das_id=(?P<id>[a-zA-Z0-9_-]+)'
	190	_TESTS = [{
	191	'url': 'https://www2.nhk.or.jp/school/movie/bangumi.cgi?das_id=D0005150191_00000',
	192	'info_dict': {
	193	'id': 'D0005150191_00003',
	194	'title': 'にているかな',
	195	'duration': 599.999,
	196	'timestamp': 1396414800,
	197
	198	'upload_date': '20140402',
	199	'ext': 'mp4',
	200
	201	'chapters': 'count:12'
	202	},
	203	'params': {
	204	# m3u8 download
	205	'skip_download': True,
	206	},
	207	}]
	208
	209	def _real_extract(self, url):
	210	program_type, video_id = self._match_valid_url(url).groups()
	211
	212	webpage = self._download_webpage(
	213	f'https://www2.nhk.or.jp/school/movie/{program_type}.cgi?das_id={video_id}', video_id)
	214
	215	# searches all variables
	216	base_values = {g.group(1): g.group(2) for g in re.finditer(r'var\s+([a-zA-Z_]+)\s=\s"([^"]+?)";', webpage)}
	217	# and programObj values too
	218	program_values = {g.group(1): g.group(3) for g in re.finditer(r'(?:program\|clip)Obj\.([a-zA-Z_]+)\s=\s(["\'])([^"]+?)\2;', webpage)}
	219	# extract all chapters
	220	chapter_durations = [parse_duration(g.group(1)) for g in re.finditer(r'chapterTime\.push\(\'([0-9:]+?)\'\);', webpage)]
	221	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)]
	222
	223	# this is how player_core.js is actually doing (!)
	224	version = base_values.get('r_version') or program_values.get('version')
	225	if version:
	226	video_id = f'{video_id.split("_")[0]}_{version}'
	227
	228	formats = self._extract_m3u8_formats(
	229	f'https://nhks-vh.akamaihd.net/i/das/{video_id[0:8]}/{video_id}_V_000.f4v/master.m3u8',
	230	video_id, ext='mp4', m3u8_id='hls')
	231	self._sort_formats(formats)
	232
	233	duration = parse_duration(base_values.get('r_duration'))
	234
	235	chapters = None
	236	if chapter_durations and chapter_titles and len(chapter_durations) == len(chapter_titles):
	237	start_time = chapter_durations
	238	end_time = chapter_durations[1:] + [duration]
	239	chapters = [{
	240	'start_time': s,
	241	'end_time': e,
	242	'title': t,
	243	} for s, e, t in zip(start_time, end_time, chapter_titles)]
	244
	245	return {
	246	'id': video_id,
	247	'title': program_values.get('name'),
	248	'duration': parse_duration(base_values.get('r_duration')),
	249	'timestamp': unified_timestamp(base_values['r_upload']),
250	'formats': formats,
251	'chapters': chapters,
252	}
253
254
255	class NhkForSchoolSubjectIE(InfoExtractor):
256	IE_DESC = 'Portal page for each school subjects, like Japanese (kokugo, 国語) or math (sansuu/suugaku or 算数・数学)'
257	KNOWN_SUBJECTS = (
258	'rika', 'syakai', 'kokugo',
259	'sansuu', 'seikatsu', 'doutoku',
260	'ongaku', 'taiiku', 'zukou',
261	'gijutsu', 'katei', 'sougou',
262	'eigo', 'tokkatsu',
263	'tokushi', 'sonota',
264	)
265	_VALID_URL = r'https?://www\.nhk\.or\.jp/school/(?P<id>%s)/?(?:[\?#].*)?$' % '\|'.join(re.escape(s) for s in KNOWN_SUBJECTS)
266
267	_TESTS = [{
268	'url': 'https://www.nhk.or.jp/school/sougou/',
269	'info_dict': {
270	'id': 'sougou',
271	'title': '総合的な学習の時間',
272	},
273	'playlist_mincount': 16,
274	}, {
275	'url': 'https://www.nhk.or.jp/school/rika/',
276	'info_dict': {
277	'id': 'rika',
278	'title': '理科',
279	},
280	'playlist_mincount': 15,
281	}]
282
283	def _real_extract(self, url):
284	subject_id = self._match_id(url)
285	webpage = self._download_webpage(url, subject_id)
286
287	return self.playlist_from_matches(
288	re.finditer(rf'href="((?:https?://www\.nhk\.or\.jp)?/school/{re.escape(subject_id)}/[^/]+/)"', webpage),
289	subject_id,
290	self._html_search_regex(r'(?s)<span\s+class="subjectName">\s<img\s[^<]+>\s*([^<]+?)</span>', webpage, 'title', fatal=False),
291	lambda g: urljoin(url, g.group(1)))
292
293
294	class NhkForSchoolProgramListIE(InfoExtractor):
295	_VALID_URL = r'https?://www\.nhk\.or\.jp/school/(?P<id>(?:%s)/[a-zA-Z0-9_-]+)' % (
296	'\|'.join(re.escape(s) for s in NhkForSchoolSubjectIE.KNOWN_SUBJECTS)
297	)
298	_TESTS = [{
299	'url': 'https://www.nhk.or.jp/school/sougou/q/',
300	'info_dict': {
301	'id': 'sougou/q',
302	'title': 'Ｑ～こどものための哲学',
303	},
304	'playlist_mincount': 20,
305	}]
306
307	def _real_extract(self, url):
308	program_id = self._match_id(url)
309
310	webpage = self._download_webpage(f'https://www.nhk.or.jp/school/{program_id}/', program_id)
311
04f3fd2c	312	title = (self._og_search_title(webpage)
	313	or self._html_extract_title(webpage)
	314	or self._html_search_regex(r'<h3>([^<]+?)とは？\s*</h3>', webpage, 'title', fatal=False))
77cc7c6e LNO	315	title = re.sub(r'\s\\|\sNHK\s+for\s+School\s*$', '', title) if title else None
	316	description = self._html_search_regex(
	317	r'(?s)<div\s+class="programDetail\s">\s<p>[^<]+</p>',
	318	webpage, 'description', fatal=False, group=0)
	319
	320	bangumi_list = self._download_json(
	321	f'https://www.nhk.or.jp/school/{program_id}/meta/program.json', program_id)
	322	# they're always bangumi
	323	bangumis = [
	324	self.url_result(f'https://www2.nhk.or.jp/school/movie/bangumi.cgi?das_id={x}')
	325	for x in traverse_obj(bangumi_list, ('part', ..., 'part-video-dasid')) or []]
	326
	327	return self.playlist_result(bangumis, program_id, title, description)