itdominator/SolarFM
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Upgrade yt_dlp and download script

Browse Source
This commit is contained in:
itdominator
2025-05-02 16:11:08 -05:00
parent 3a2e8eeb08
commit d68d9ce4f9
1194 changed files with 60099 additions and 44436 deletions
Download Patch File Download Diff File Expand all files Collapse all files

611
plugins/youtube_download/yt_dlp/extractor/nhk.py
View File

@@ -3,10 +3,15 @@ import re
from .common import InfoExtractor
from ..utils import (
ExtractorError,
clean_html,
filter_dict,
get_element_by_class,
int_or_none,
join_nonempty,
parse_duration,
remove_end,
traverse_obj,
try_call,
unescapeHTML,
unified_timestamp,
url_or_none,
@@ -16,8 +21,7 @@ from ..utils import (
class NhkBaseIE(InfoExtractor):
_API_URL_TEMPLATE = 'https://nwapi.nhk.jp/nhkworld/%sod%slist/v7b/%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)/'
_BASE_URL_REGEX = r'https?://www3\.nhk\.or\.jp/nhkworld/(?P<lang>[a-z]{2})/'
def _call_api(self, m_id, lang, is_video, is_episode, is_clip):
return self._download_json(
@@ -28,21 +32,71 @@ class NhkBaseIE(InfoExtractor):
m_id, lang, '/all' if is_video else ''),
m_id, query={'apikey': 'EJfK8jdS57GqlupFgAfAAwr573q01y6k'})['data']['episodes'] or []
def _get_api_info(self, refresh=True):
if not refresh:
return self.cache.load('nhk', 'api_info')
self.cache.store('nhk', 'api_info', {})
movie_player_js = self._download_webpage(
'https://movie-a.nhk.or.jp/world/player/js/movie-player.js', None,
note='Downloading stream API information')
api_info = {
'url': self._search_regex(
r'prod:[^;]+\bapiUrl:\s*[\'"]([^\'"]+)[\'"]', movie_player_js, None, 'stream API url'),
'token': self._search_regex(
r'prod:[^;]+\btoken:\s*[\'"]([^\'"]+)[\'"]', movie_player_js, None, 'stream API token'),
}
self.cache.store('nhk', 'api_info', api_info)
return api_info
def _extract_stream_info(self, vod_id):
for refresh in (False, True):
api_info = self._get_api_info(refresh)
if not api_info:
continue
api_url = api_info.pop('url')
meta = traverse_obj(
self._download_json(
api_url, vod_id, 'Downloading stream url info', fatal=False, query={
**api_info,
'type': 'json',
'optional_id': vod_id,
'active_flg': 1,
}), ('meta', 0))
stream_url = traverse_obj(
meta, ('movie_url', ('mb_auto', 'auto_sp', 'auto_pc'), {url_or_none}), get_all=False)
if stream_url:
formats, subtitles = self._extract_m3u8_formats_and_subtitles(stream_url, vod_id)
return {
**traverse_obj(meta, {
'duration': ('duration', {int_or_none}),
'timestamp': ('publication_date', {unified_timestamp}),
'release_timestamp': ('insert_date', {unified_timestamp}),
'modified_timestamp': ('update_date', {unified_timestamp}),
}),
'formats': formats,
'subtitles': subtitles,
}
raise ExtractorError('Unable to extract stream url')
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 len(episode_id) == 7:
lang, m_type, episode_id = NhkVodIE._match_valid_url(url).group('lang', 'type', 'id')
is_video = m_type != 'audio'
if is_video:
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)
return clean_html(episode.get(key + '_clean') or episode.get(key))
title = get_clean_field('sub_title')
series = get_clean_field('title')
thumbnails = []
@@ -51,33 +105,44 @@ class NhkBaseIE(InfoExtractor):
if not img_path:
continue
thumbnails.append({
'id': '%dp' % h,
'id': f'{h}p',
'height': h,
'width': w,
'url': 'https://www3.nhk.or.jp' + img_path,
})
episode_name = title
if series and title:
title = f'{series} - {title}'
elif series and not title:
title = series
series = None
episode_name = None
else: # title, no series
episode_name = None
info = {
'id': episode_id + '-' + lang,
'title': '%s - %s' % (series, title) if series and title else title,
'title': title,
'description': get_clean_field('description'),
'thumbnails': thumbnails,
'series': series,
'episode': title,
'episode': episode_name,
}
if is_video:
vod_id = episode['vod_id']
info.update({
'_type': 'url_transparent',
'ie_key': 'Piksel',
'url': 'https://movie-s.nhk.or.jp/v/refid/nhkworld/prefid/' + vod_id,
**self._extract_stream_info(vod_id),
'id': vod_id,
})
else:
if fetch_episode:
audio_path = episode['audio']['audio']
# From https://www3.nhk.or.jp/nhkworld/common/player/radio/inline/rod.html
audio_path = remove_end(episode['audio']['audio'], '.m4a')
info['formats'] = self._extract_m3u8_formats(
'https://nhkworld-vh.akamaihd.net/i%s/master.m3u8' % audio_path,
f'{urljoin("https://vod-stream.nhk.jp", audio_path)}/index.m3u8',
episode_id, 'm4a', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False)
for f in info['formats']:
@@ -92,48 +157,63 @@ class NhkBaseIE(InfoExtractor):
class NhkVodIE(NhkBaseIE):
# the 7-character IDs can have alphabetic chars too: assume [a-z] rather than just [a-f], eg
_VALID_URL = r'%s%s(?P<id>[0-9a-z]{7}|[^/]+?-\d{8}-[0-9a-z]+)' % (NhkBaseIE._BASE_URL_REGEX, NhkBaseIE._TYPE_REGEX)
_VALID_URL = [
rf'{NhkBaseIE._BASE_URL_REGEX}shows/(?:(?P<type>video)/)?(?P<id>\d{{4}}[\da-z]\d+)/?(?:$|[?#])',
rf'{NhkBaseIE._BASE_URL_REGEX}(?:ondemand|shows)/(?P<type>audio)/(?P<id>[^/?#]+?-\d{{8}}-[\da-z]+)',
rf'{NhkBaseIE._BASE_URL_REGEX}ondemand/(?P<type>video)/(?P<id>\d{{4}}[\da-z]\d+)', # deprecated
]
# Content available only for a limited period of time. Visit
# https://www3.nhk.or.jp/nhkworld/en/ondemand/ for working samples.
_TESTS = [{
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/2061601/',
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/2049126/',
'info_dict': {
'id': 'yd8322ch',
'id': 'nw_vod_v_en_2049_126_20230413233000_01_1681398302',
'ext': 'mp4',
'description': 'md5:109c8b05d67a62d0592f2b445d2cd898',
'title': 'GRAND SUMO Highlights - [Recap] May Tournament Day 1 (Opening Day)',
'upload_date': '20230514',
'timestamp': 1684083791,
'series': 'GRAND SUMO Highlights',
'episode': '[Recap] May Tournament Day 1 (Opening Day)',
'thumbnail': 'https://mz-edge.stream.co.jp/thumbs/aid/t1684084443/4028649.jpg?w=1920&h=1080',
'title': 'Japan Railway Journal - The Tohoku Shinkansen: Full Speed Ahead',
'description': 'md5:49f7c5b206e03868a2fdf0d0814b92f6',
'thumbnail': r're:https://.+/.+\.jpg',
'episode': 'The Tohoku Shinkansen: Full Speed Ahead',
'series': 'Japan Railway Journal',
'modified_timestamp': 1707217907,
'timestamp': 1681428600,
'release_timestamp': 1693883728,
'duration': 1679,
'upload_date': '20230413',
'modified_date': '20240206',
'release_date': '20230905',
},
}, {
# video clip
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/9999011/',
'md5': '7a90abcfe610ec22a6bfe15bd46b30ca',
'md5': '153c3016dfd252ba09726588149cf0e7',
'info_dict': {
'id': 'a95j5iza',
'id': 'lpZXIwaDE6_Z-976CPsFdxyICyWUzlT5',
'ext': 'mp4',
'title': "Dining with the Chef - Chef Saito's Family recipe: MENCHI-KATSU",
'title': 'Dining with the Chef - Chef Saito\'s Family recipe: MENCHI-KATSU',
'description': 'md5:5aee4a9f9d81c26281862382103b0ea5',
'timestamp': 1565965194,
'upload_date': '20190816',
'thumbnail': 'https://mz-edge.stream.co.jp/thumbs/aid/t1567086278/3715195.jpg?w=1920&h=1080',
'thumbnail': r're:https://.+/.+\.jpg',
'series': 'Dining with the Chef',
'episode': 'Chef Saito\'s Family recipe: MENCHI-KATSU',
'duration': 148,
'upload_date': '20190816',
'release_date': '20230902',
'release_timestamp': 1693619292,
'modified_timestamp': 1707217907,
'modified_date': '20240206',
'timestamp': 1565997540,
},
}, {
# audio clip
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/r_inventions-20201104-1/',
# radio
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/livinginjapan-20231001-1/',
'info_dict': {
'id': 'r_inventions-20201104-1-en',
'id': 'livinginjapan-20231001-1-en',
'ext': 'm4a',
'title': "Japan's Top Inventions - Miniature Video Cameras",
'description': 'md5:07ea722bdbbb4936fdd360b6a480c25b',
'title': 'Living in Japan - Tips for Travelers to Japan / Ramen Vending Machines',
'series': 'Living in Japan',
'description': 'md5:0a0e2077d8f07a03071e990a6f51bfab',
'thumbnail': r're:https://.+/.+\.jpg',
'episode': 'Tips for Travelers to Japan / Ramen Vending Machines',
},
'skip': '404 Not Found',
}, {
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/2015173/',
'only_matching': True,
@@ -159,6 +239,119 @@ class NhkVodIE(NhkBaseIE):
'timestamp': 1623722008,
},
'skip': '404 Not Found',
}, {
# japanese-language, longer id than english
'url': 'https://www3.nhk.or.jp/nhkworld/ja/ondemand/video/0020271111/',
'info_dict': {
'id': 'nw_ja_v_jvod_ohayou_20231008',
'ext': 'mp4',
'title': 'おはよう日本（7時台） - 10月8日放送',
'series': 'おはよう日本（7時台）',
'episode': '10月8日放送',
'thumbnail': r're:https://.+/.+\.jpg',
'description': 'md5:9c1d6cbeadb827b955b20e99ab920ff0',
},
'skip': 'expires 2023-10-15',
}, {
# a one-off (single-episode series). title from the api is just '<p></p>'
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/3004952/',
'info_dict': {
'id': 'nw_vod_v_en_3004_952_20230723091000_01_1690074552',
'ext': 'mp4',
'title': 'Barakan Discovers - AMAMI OSHIMA: Isson\'s Treasure Isla',
'description': 'md5:5db620c46a0698451cc59add8816b797',
'thumbnail': r're:https://.+/.+\.jpg',
'release_date': '20230905',
'timestamp': 1690103400,
'duration': 2939,
'release_timestamp': 1693898699,
'upload_date': '20230723',
'modified_timestamp': 1707217907,
'modified_date': '20240206',
'episode': 'AMAMI OSHIMA: Isson\'s Treasure Isla',
'series': 'Barakan Discovers',
},
}, {
# /ondemand/video/ url with alphabetical character in 5th position of id
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/9999a07/',
'info_dict': {
'id': 'nw_c_en_9999-a07',
'ext': 'mp4',
'episode': 'Mini-Dramas on SDGs: Ep 1 Close the Gender Gap [Director\'s Cut]',
'series': 'Mini-Dramas on SDGs',
'modified_date': '20240206',
'title': 'Mini-Dramas on SDGs - Mini-Dramas on SDGs: Ep 1 Close the Gender Gap [Director\'s Cut]',
'description': 'md5:3f9dcb4db22fceb675d90448a040d3f6',
'timestamp': 1621962360,
'duration': 189,
'release_date': '20230903',
'modified_timestamp': 1707217907,
'upload_date': '20210525',
'thumbnail': r're:https://.+/.+\.jpg',
'release_timestamp': 1693713487,
},
}, {
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/9999d17/',
'info_dict': {
'id': 'nw_c_en_9999-d17',
'ext': 'mp4',
'title': 'Flowers of snow blossom - The 72 Pentads of Yamato',
'description': 'Today’s focus: Snow',
'release_timestamp': 1693792402,
'release_date': '20230904',
'upload_date': '20220128',
'timestamp': 1643370960,
'thumbnail': r're:https://.+/.+\.jpg',
'duration': 136,
'series': '',
'modified_date': '20240206',
'modified_timestamp': 1707217907,
},
}, {
# new /shows/ url format
'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/2032307/',
'info_dict': {
'id': 'nw_vod_v_en_2032_307_20240321113000_01_1710990282',
'ext': 'mp4',
'title': 'Japanology Plus - 20th Anniversary Special Part 1',
'description': 'md5:817d41fc8e54339ad2a916161ea24faf',
'episode': '20th Anniversary Special Part 1',
'series': 'Japanology Plus',
'thumbnail': r're:https://.+/.+\.jpg',
'duration': 1680,
'timestamp': 1711020600,
'upload_date': '20240321',
'release_timestamp': 1711022683,
'release_date': '20240321',
'modified_timestamp': 1711031012,
'modified_date': '20240321',
},
}, {
'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/3020025/',
'info_dict': {
'id': 'nw_vod_v_en_3020_025_20230325144000_01_1679723944',
'ext': 'mp4',
'title': '100 Ideas to Save the World - Working Styles Evolve',
'description': 'md5:9e6c7778eaaf4f7b4af83569649f84d9',
'episode': 'Working Styles Evolve',
'series': '100 Ideas to Save the World',
'thumbnail': r're:https://.+/.+\.jpg',
'duration': 899,
'upload_date': '20230325',
'timestamp': 1679755200,
'release_date': '20230905',
'release_timestamp': 1693880540,
'modified_date': '20240206',
'modified_timestamp': 1707217907,
},
}, {
# new /shows/audio/ url format
'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/audio/livinginjapan-20231001-1/',
'only_matching': True,
}, {
# valid url even if can't be found in wild; support needed for clip entries extraction
'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/9999o80/',
'only_matching': True,
}]
def _real_extract(self, url):
@@ -166,58 +359,91 @@ class NhkVodIE(NhkBaseIE):
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)
_VALID_URL = rf'''(?x)
{NhkBaseIE._BASE_URL_REGEX}(?:shows|tv)/
(?:(?P<type>audio)/programs/)?(?P<id>\w+)/?
(?:\?(?:[^#]+&)?type=(?P<episode_type>clip|(?:radio|tv)Episode))?'''
_TESTS = [{
# video program episodes
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/sumo',
'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/sumo/',
'info_dict': {
'id': 'sumo',
'title': 'GRAND SUMO Highlights',
'description': 'md5:fc20d02dc6ce85e4b72e0273aa52fdbf',
},
'playlist_mincount': 12,
'playlist_mincount': 1,
}, {
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/japanrailway',
'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/japanrailway/',
'info_dict': {
'id': 'japanrailway',
'title': 'Japan Railway Journal',
'description': 'md5:ea39d93af7d05835baadf10d1aae0e3f',
},
'playlist_mincount': 12,
}, {
# video program clips
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/japanrailway/?type=clip',
'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/japanrailway/?type=clip',
'info_dict': {
'id': 'japanrailway',
'title': 'Japan Railway Journal',
'description': 'md5:ea39d93af7d05835baadf10d1aae0e3f',
},
'playlist_mincount': 5,
}, {
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/10yearshayaomiyazaki/',
'only_matching': True,
'playlist_mincount': 12,
}, {
# audio program
'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/audio/listener/',
'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/audio/programs/livinginjapan/',
'info_dict': {
'id': 'livinginjapan',
'title': 'Living in Japan',
'description': 'md5:665bb36ec2a12c5a7f598ee713fc2b54',
},
'playlist_mincount': 12,
}, {
# /tv/ program url
'url': 'https://www3.nhk.or.jp/nhkworld/en/tv/designtalksplus/',
'info_dict': {
'id': 'designtalksplus',
'title': 'DESIGN TALKS plus',
'description': 'md5:47b3b3a9f10d4ac7b33b53b70a7d2837',
},
'playlist_mincount': 20,
}, {
'url': 'https://www3.nhk.or.jp/nhkworld/en/shows/10yearshayaomiyazaki/',
'only_matching': True,
}]
@classmethod
def suitable(cls, url):
return False if NhkVodIE.suitable(url) else super().suitable(url)
def _extract_meta_from_class_elements(self, class_values, html):
for class_value in class_values:
if value := clean_html(get_element_by_class(class_value, html)):
return value
def _real_extract(self, url):
lang, m_type, program_id, episode_type = self._match_valid_url(url).groups()
lang, m_type, program_id, episode_type = self._match_valid_url(url).group('lang', 'type', 'id', 'episode_type')
episodes = self._call_api(
program_id, lang, m_type == 'video', False, episode_type == 'clip')
program_id, lang, m_type != 'audio', 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))
def entries():
for episode in episodes:
if episode_path := episode.get('url'):
yield self._extract_episode_info(urljoin(url, episode_path), episode)
program_title = None
if entries:
program_title = entries[0].get('series')
html = self._download_webpage(url, program_id)
program_title = self._extract_meta_from_class_elements([
'p-programDetail__title', # /ondemand/program/
'pProgramHero__logoText', # /shows/
'tAudioProgramMain__title', # /shows/audio/programs/
'p-program-name'], html) # /tv/
program_description = self._extract_meta_from_class_elements([
'p-programDetail__text', # /ondemand/program/
'pProgramHero__description', # /shows/
'tAudioProgramMain__info', # /shows/audio/programs/
'p-program-description'], html) # /tv/
return self.playlist_result(entries, program_id, program_title)
return self.playlist_result(entries(), program_id, program_title, program_description)
class NhkForSchoolBangumiIE(InfoExtractor):
@@ -233,7 +459,7 @@ class NhkForSchoolBangumiIE(InfoExtractor):
'upload_date': '20140402',
'ext': 'mp4',
'chapters': 'count:12'
'chapters': 'count:12',
},
'params': {
# m3u8 download
@@ -296,7 +522,8 @@ class NhkForSchoolSubjectIE(InfoExtractor):
'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)
_VALID_URL = r'https?://www\.nhk\.or\.jp/school/(?P<id>{})/?(?:[\?#].*)?$'.format(
'|'.join(re.escape(s) for s in KNOWN_SUBJECTS))
_TESTS = [{
'url': 'https://www.nhk.or.jp/school/sougou/',
@@ -326,9 +553,8 @@ class NhkForSchoolSubjectIE(InfoExtractor):
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)
)
_VALID_URL = r'https?://www\.nhk\.or\.jp/school/(?P<id>(?:{})/[a-zA-Z0-9_-]+)'.format(
'|'.join(re.escape(s) for s in NhkForSchoolSubjectIE.KNOWN_SUBJECTS))
_TESTS = [{
'url': 'https://www.nhk.or.jp/school/sougou/q/',
'info_dict': {
@@ -365,21 +591,22 @@ class NhkRadiruIE(InfoExtractor):
IE_DESC = 'NHK らじる (Radiru/Rajiru)'
_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]+))?'
_TESTS = [{
'url': 'https://www.nhk.or.jp/radio/player/ondemand.html?p=0449_01_3853544',
'skip': 'Episode expired on 2023-04-16',
'url': 'https://www.nhk.or.jp/radio/player/ondemand.html?p=0449_01_4003239',
'skip': 'Episode expired on 2024-06-09',
'info_dict': {
'channel': 'NHK-FM',
'description': 'md5:94b08bdeadde81a97df4ec882acce3e9',
'title': 'ジャズ・トゥナイト　ジャズ「Ｎｉｇｈｔ　ａｎｄ　Ｄａｙ」特集',
'id': '0449_01_4003239',
'ext': 'm4a',
'id': '0449_01_3853544',
'uploader': 'NHK FM 東京',
'description': 'md5:ad05f3c3f3f6e99b2e69f9b5e49551dc',
'series': 'ジャズ・トゥナイト',
'channel': 'NHK FM 東京',
'thumbnail': 'https://www.nhk.or.jp/prog/img/449/g449.jpg',
'timestamp': 1680969600,
'title': 'ジャズ・トゥナイト　ＮＥＷジャズ特集',
'upload_date': '20230408',
'release_timestamp': 1680962400,
'release_date': '20230408',
'was_live': True,
'upload_date': '20240601',
'series_id': '0449_01',
'release_date': '20240601',
'timestamp': 1717257600,
'release_timestamp': 1717250400,
},
}, {
# playlist, airs every weekday so it should _hopefully_ be okay forever
@@ -388,46 +615,143 @@ class NhkRadiruIE(InfoExtractor):
'id': '0458_01',
'title': 'ベストオブクラシック',
'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。',
'channel': 'NHK-FM',
'thumbnail': 'https://www.nhk.or.jp/prog/img/458/g458.jpg',
'series_id': '0458_01',
'uploader': 'NHK FM',
'channel': 'NHK FM',
'series': 'ベストオブクラシック',
},
'playlist_mincount': 3,
}, {
# one with letters in the id
'url': 'https://www.nhk.or.jp/radio/player/ondemand.html?p=F300_06_3738470',
'note': 'Expires on 2024-03-31',
'url': 'https://www.nhk.or.jp/radio/player/ondemand.html?p=F683_01_3910688',
'note': 'Expires on 2025-03-31',
'info_dict': {
'id': 'F300_06_3738470',
'id': 'F683_01_3910688',
'ext': 'm4a',
'title': '有島武郎「一房のぶどう」',
'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）',
'channel': 'NHKラジオ第1、NHK-FM',
'timestamp': 1635757200,
'thumbnail': 'https://www.nhk.or.jp/radioondemand/json/F300/img/corner/box_109_thumbnail.jpg',
'release_date': '20161207',
'series': 'らじる文庫 by ラジオ深夜便 ',
'release_timestamp': 1481126700,
'upload_date': '20211101',
}
'title': '夏目漱石「文鳥」第1回',
'series': '【らじる文庫】夏目漱石「文鳥」（全4回）',
'series_id': 'F683_01',
'description': '朗読：浅井理アナウンサー',
'thumbnail': 'https://www.nhk.or.jp/radioondemand/json/F683/img/roudoku_05_rod_640.jpg',
'upload_date': '20240106',
'release_date': '20240106',
'uploader': 'NHK R1',
'release_timestamp': 1704511800,
'channel': 'NHK R1',
'timestamp': 1704512700,
},
'expected_warnings': ['Unable to download JSON metadata',
'Failed to get extended metadata. API returned Error 1: Invalid parameters'],
}, {
# news
'url': 'https://www.nhk.or.jp/radio/player/ondemand.html?p=F261_01_3855109',
'skip': 'Expires on 2023-04-17',
'url': 'https://www.nhk.or.jp/radio/player/ondemand.html?p=F261_01_4012173',
'info_dict': {
'id': 'F261_01_3855109',
'id': 'F261_01_4012173',
'ext': 'm4a',
'channel': 'NHKラジオ第1',
'timestamp': 1681635900,
'release_date': '20230416',
'uploader': 'NHKラジオ第1',
'series': 'NHKラジオニュース',
'title': '午後６時のNHKニュース',
'title': '午前０時のNHKニュース',
'thumbnail': 'https://www.nhk.or.jp/radioondemand/json/F261/img/RADIONEWS_640.jpg',
'upload_date': '20230416',
'release_timestamp': 1681635600,
'release_timestamp': 1718290800,
'release_date': '20240613',
'timestamp': 1718291400,
'upload_date': '20240613',
},
}, {
# fallback when extended metadata fails
'url': 'https://www.nhk.or.jp/radio/player/ondemand.html?p=2834_01_4009298',
'skip': 'Expires on 2024-06-07',
'info_dict': {
'id': '2834_01_4009298',
'title': 'まち☆キラ！開成町特集',
'ext': 'm4a',
'release_date': '20240531',
'upload_date': '20240531',
'series': 'はま☆キラ！',
'thumbnail': 'https://www.nhk.or.jp/prog/img/2834/g2834.jpg',
'channel': 'NHK R1,FM',
'description': '',
'timestamp': 1717123800,
'uploader': 'NHK R1,FM',
'release_timestamp': 1717120800,
'series_id': '2834_01',
},
'expected_warnings': ['Failed to get extended metadata. API returned empty list.'],
}]
def _extract_episode_info(self, headline, programme_id, series_meta):
_API_URL_TMPL = None
def _extract_extended_metadata(self, episode_id, aa_vinfo):
service, _, area = traverse_obj(aa_vinfo, (2, {str}, {lambda x: (x or '').partition(',')}))
detail_url = try_call(
lambda: self._API_URL_TMPL.format(area=area, service=service, dateid=aa_vinfo[3]))
if not detail_url:
return {}
response = self._download_json(
detail_url, episode_id, 'Downloading extended metadata',
'Failed to download extended metadata', fatal=False, expected_status=400)
if not response:
return {}
if error := traverse_obj(response, ('error', {dict})):
self.report_warning(
'Failed to get extended metadata. API returned '
f'Error {join_nonempty("code", "message", from_dict=error, delim=": ")}')
return {}
full_meta = traverse_obj(response, ('list', service, 0, {dict}))
if not full_meta:
self.report_warning('Failed to get extended metadata. API returned empty list.')
return {}
station = ' '.join(traverse_obj(full_meta, (('service', 'area'), 'name', {str}))) or None
thumbnails = [{
'id': str(id_),
'preference': 1 if id_.startswith('thumbnail') else -2 if id_.startswith('logo') else -1,
**traverse_obj(thumb, {
'url': 'url',
'width': ('width', {int_or_none}),
'height': ('height', {int_or_none}),
}),
} for id_, thumb in traverse_obj(full_meta, ('images', {dict.items}, lambda _, v: v[1]['url']))]
return filter_dict({
'channel': station,
'uploader': station,
'description': join_nonempty(
'subtitle', 'content', 'act', 'music', delim='\n\n', from_dict=full_meta),
'thumbnails': thumbnails,
**traverse_obj(full_meta, {
'title': ('title', {str}),
'timestamp': ('end_time', {unified_timestamp}),
'release_timestamp': ('start_time', {unified_timestamp}),
}),
})
def _extract_episode_info(self, episode, programme_id, series_meta):
episode_id = f'{programme_id}_{episode["id"]}'
aa_vinfo = traverse_obj(episode, ('aa_contents_id', {lambda x: x.split(';')}))
extended_metadata = self._extract_extended_metadata(episode_id, aa_vinfo)
fallback_start_time, _, fallback_end_time = traverse_obj(
aa_vinfo, (4, {str}, {lambda x: (x or '').partition('_')}))
return {
**series_meta,
'id': episode_id,
'formats': self._extract_m3u8_formats(episode.get('stream_url'), episode_id, fatal=False),
'container': 'm4a_dash', # force fixup, AAC-only HLS
'was_live': True,
'title': episode.get('program_title'),
'description': episode.get('program_sub_title'), # fallback
'timestamp': unified_timestamp(fallback_end_time),
'release_timestamp': unified_timestamp(fallback_start_time),
**extended_metadata,
}
def _extract_news_info(self, headline, programme_id, series_meta):
episode_id = f'{programme_id}_{headline["headline_id"]}'
episode = traverse_obj(headline, ('file_list', 0, {dict}))
@@ -440,42 +764,76 @@ class NhkRadiruIE(InfoExtractor):
'series': series_meta.get('title'),
'thumbnail': url_or_none(headline.get('headline_image')) or series_meta.get('thumbnail'),
**traverse_obj(episode, {
'title': 'file_title',
'description': 'file_title_sub',
'title': ('file_title', {str}),
'description': ('file_title_sub', {str}),
'timestamp': ('open_time', {unified_timestamp}),
'release_timestamp': ('aa_vinfo4', {lambda x: x.split('_')[0]}, {unified_timestamp}),
}),
}
def _real_initialize(self):
if self._API_URL_TMPL:
return
api_config = self._download_xml(
'https://www.nhk.or.jp/radio/config/config_web.xml', None, 'Downloading API config', fatal=False)
NhkRadiruIE._API_URL_TMPL = try_call(lambda: f'https:{api_config.find(".//url_program_detail").text}')
def _real_extract(self, url):
site_id, corner_id, headline_id = self._match_valid_url(url).group('site', 'corner', 'headline')
programme_id = f'{site_id}_{corner_id}'
if site_id == 'F261':
json_url = 'https://www.nhk.or.jp/s-media/news/news-site/list/v1/all.json'
else:
json_url = f'https://www.nhk.or.jp/radioondemand/json/{site_id}/bangumi_{programme_id}.json'
if site_id == 'F261': # XXX: News programmes use old API (for now?)
meta = self._download_json(
'https://www.nhk.or.jp/s-media/news/news-site/list/v1/all.json', programme_id)['main']
series_meta = traverse_obj(meta, {
'title': ('program_name', {str}),
'channel': ('media_name', {str}),
'uploader': ('media_name', {str}),
'thumbnail': (('thumbnail_c', 'thumbnail_p'), {url_or_none}),
}, get_all=False)
meta = self._download_json(json_url, programme_id)['main']
if headline_id:
headline = traverse_obj(
meta, ('detail_list', lambda _, v: v['headline_id'] == headline_id, any))
if not headline:
raise ExtractorError('Content not found; it has most likely expired', expected=True)
return self._extract_news_info(headline, programme_id, series_meta)
series_meta = traverse_obj(meta, {
'title': 'program_name',
'channel': 'media_name',
'thumbnail': (('thumbnail_c', 'thumbnail_p'), {url_or_none}),
}, get_all=False)
def news_entries():
for headline in traverse_obj(meta, ('detail_list', ..., {dict})):
yield self._extract_news_info(headline, programme_id, series_meta)
return self.playlist_result(
news_entries(), programme_id, description=meta.get('site_detail'), **series_meta)
meta = self._download_json(
'https://www.nhk.or.jp/radio-api/app/v1/web/ondemand/series', programme_id, query={
'site_id': site_id,
'corner_site_id': corner_id,
})
fallback_station = join_nonempty('NHK', traverse_obj(meta, ('radio_broadcast', {str})), delim=' ')
series_meta = {
'series': join_nonempty('title', 'corner_name', delim=' ', from_dict=meta),
'series_id': programme_id,
'thumbnail': traverse_obj(meta, ('thumbnail_url', {url_or_none})),
'channel': fallback_station,
'uploader': fallback_station,
}
if headline_id:
return self._extract_episode_info(
traverse_obj(meta, (
'detail_list', lambda _, v: v['headline_id'] == headline_id), get_all=False),
programme_id, series_meta)
episode = traverse_obj(meta, ('episodes', lambda _, v: v['id'] == int(headline_id), any))
if not episode:
raise ExtractorError('Content not found; it has most likely expired', expected=True)
return self._extract_episode_info(episode, programme_id, series_meta)
def entries():
for headline in traverse_obj(meta, ('detail_list', ..., {dict})):
yield self._extract_episode_info(headline, programme_id, series_meta)
for episode in traverse_obj(meta, ('episodes', ..., {dict})):
yield self._extract_episode_info(episode, programme_id, series_meta)
return self.playlist_result(
entries(), programme_id, playlist_description=meta.get('site_detail'), **series_meta)
entries(), programme_id, title=series_meta.get('series'),
description=meta.get('series_description'), **series_meta)
class NhkRadioNewsPageIE(InfoExtractor):
@@ -489,8 +847,9 @@ class NhkRadioNewsPageIE(InfoExtractor):
'thumbnail': 'https://www.nhk.or.jp/radioondemand/json/F261/img/RADIONEWS_640.jpg',
'description': 'md5:bf2c5b397e44bc7eb26de98d8f15d79d',
'channel': 'NHKラジオ第1',
'uploader': 'NHKラジオ第1',
'title': 'NHKラジオニュース',
}
},
}]
def _real_extract(self, url):
@@ -532,7 +891,7 @@ class NhkRadiruLiveIE(InfoExtractor):
'ext': 'm4a',
'thumbnail': 'https://www.nhk.or.jp/common/img/media/fm-200x200.png',
'live_status': 'is_live',
}
},
}]
_NOA_STATION_IDS = {'r1': 'n1', 'r2': 'n2', 'fm': 'n3'}
@@ -546,16 +905,16 @@ class NhkRadiruLiveIE(InfoExtractor):
data = config.find(f'.//data//area[.="{area}"]/..')
if not data:
raise ExtractorError('Invalid area. Valid areas are: %s' % ', '.join(
[i.text for i in config.findall('.//data//area')]), expected=True)
raise ExtractorError('Invalid area. Valid areas are: {}'.format(', '.join(
[i.text for i in config.findall('.//data//area')])), expected=True)
noa_info = self._download_json(
f'https:{config.find(".//url_program_noa").text}'.format(area=data.find('areakey').text),
station, note=f'Downloading {area} station metadata')
station, note=f'Downloading {area} station metadata', fatal=False)
present_info = traverse_obj(noa_info, ('nowonair_list', self._NOA_STATION_IDS.get(station), 'present'))
return {
'title': ' '.join(traverse_obj(present_info, (('service', 'area',), 'name', {str}))),
'title': ' '.join(traverse_obj(present_info, (('service', 'area'), 'name', {str}))),
'id': join_nonempty(station, area),
'thumbnails': traverse_obj(present_info, ('service', 'images', ..., {
'url': 'url',