diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index a487899b4..fb0277063 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -808,6 +808,7 @@ from .nrk import ( NRKTVSeasonIE, NRKTVSeriesIE, ) +from .ntvcojp import NTVCoJpCUIE from .ntvde import NTVDeIE from .ntvru import NTVRuIE from .nytimes import ( diff --git a/youtube_dl/extractor/nhk.py b/youtube_dl/extractor/nhk.py index d4acbcc3e..241412f98 100644 --- a/youtube_dl/extractor/nhk.py +++ b/youtube_dl/extractor/nhk.py @@ -1,54 +1,81 @@ from __future__ import unicode_literals +import re + from .common import InfoExtractor -from ..utils import ExtractorError class NhkVodIE(InfoExtractor): - _VALID_URL = r'https?://www3\.nhk\.or\.jp/nhkworld/en/(?:vod|ondemand)/(?P[^/]+/[^/?#&]+)' + _VALID_URL = r'https?://www3\.nhk\.or\.jp/nhkworld/(?P[a-z]{2})/ondemand/(?Pvideo|audio)/(?P\d{7}|[a-z]+-\d{8}-\d+)' + # Content available only for a limited period of time. Visit + # https://www3.nhk.or.jp/nhkworld/en/ondemand/ for working samples. _TESTS = [{ - # Videos available only for a limited period of time. Visit - # http://www3.nhk.or.jp/nhkworld/en/vod/ for working samples. - 'url': 'http://www3.nhk.or.jp/nhkworld/en/vod/tokyofashion/20160815', - 'info_dict': { - 'id': 'A1bnNiNTE6nY3jLllS-BIISfcC_PpvF5', - 'ext': 'flv', - 'title': 'TOKYO FASHION EXPRESS - The Kimono as Global Fashion', - 'description': 'md5:db338ee6ce8204f415b754782f819824', - 'series': 'TOKYO FASHION EXPRESS', - 'episode': 'The Kimono as Global Fashion', - }, - 'skip': 'Videos available only for a limited period of time', - }, { '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, }] - _API_URL = 'http://api.nhk.or.jp/nhkworld/vodesdlist/v1/all/all/all.json?apikey=EJfK8jdS57GqlupFgAfAAwr573q01y6k' + _API_URL_TEMPLATE = 'https://api.nhk.or.jp/nhkworld/%sodesdlist/v7/episode/%s/%s/all%s.json' def _real_extract(self, url): - video_id = self._match_id(url) - - data = self._download_json(self._API_URL, video_id) - - try: - episode = next( - e for e in data['data']['episodes'] - if e.get('url') and video_id in e['url']) - except StopIteration: - raise ExtractorError('Unable to find episode') - - embed_code = episode['vod_id'] + lang, m_type, episode_id = re.match(self._VALID_URL, url).groups() + if episode_id.isdigit(): + episode_id = episode_id[:4] + '-' + episode_id[4:] + is_video = m_type == 'video' + episode = self._download_json( + self._API_URL_TEMPLATE % ('v' if is_video else 'r', episode_id, lang, '/all' if is_video else ''), + episode_id, query={'apikey': 'EJfK8jdS57GqlupFgAfAAwr573q01y6k'})['data']['episodes'][0] title = episode.get('sub_title_clean') or episode['sub_title'] - description = episode.get('description_clean') or episode.get('description') - series = episode.get('title_clean') or episode.get('title') - return { - '_type': 'url_transparent', - 'ie_key': 'Ooyala', - 'url': 'ooyala:%s' % embed_code, + 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': description, + 'description': get_clean_field('description'), + 'thumbnails': thumbnails, 'series': series, 'episode': title, } + if is_video: + info.update({ + '_type': 'url_transparent', + 'ie_key': 'Ooyala', + 'url': 'ooyala:' + episode['vod_id'], + }) + else: + audio = episode['audio'] + audio_path = audio['audio'] + info['formats'] = self._extract_m3u8_formats( + 'https://nhks-vh.akamaihd.net/i%s/master.m3u8' % audio_path, + episode_id, 'm4a', m3u8_id='hls', fatal=False) + for proto in ('rtmpt', 'rtmp'): + info['formats'].append({ + 'ext': 'flv', + 'format_id': proto, + 'url': '%s://flv.nhk.or.jp/ondemand/mp4:flv%s' % (proto, audio_path), + 'vcodec': 'none', + }) + for f in info['formats']: + f['language'] = lang + return info diff --git a/youtube_dl/extractor/ntvcojp.py b/youtube_dl/extractor/ntvcojp.py new file mode 100644 index 000000000..0c8221b22 --- /dev/null +++ b/youtube_dl/extractor/ntvcojp.py @@ -0,0 +1,49 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + js_to_json, + smuggle_url, +) + + +class NTVCoJpCUIE(InfoExtractor): + IE_NAME = 'cu.ntv.co.jp' + IE_DESC = 'Nippon Television Network' + _VALID_URL = r'https?://cu\.ntv\.co\.jp/(?!program)(?P[^/?&#]+)' + _TEST = { + 'url': 'https://cu.ntv.co.jp/televiva-chill-gohan_181031/', + 'info_dict': { + 'id': '5978891207001', + 'ext': 'mp4', + 'title': '桜エビと炒り卵がポイント! 「中華風 エビチリおにぎり」──『美虎』五十嵐美幸', + 'upload_date': '20181213', + 'description': 'md5:211b52f4fd60f3e0e72b68b0c6ba52a9', + 'uploader_id': '3855502814001', + 'timestamp': 1544669941, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + } + BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s' + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + player_config = self._parse_json(self._search_regex( + r'(?s)PLAYER_CONFIG\s*=\s*({.+?})', + webpage, 'player config'), display_id, js_to_json) + video_id = player_config['videoId'] + account_id = player_config.get('account') or '3855502814001' + return { + '_type': 'url_transparent', + 'id': video_id, + 'display_id': display_id, + 'title': self._search_regex(r']+class="title"[^>]*>([^<]+)', webpage, 'title').strip(), + 'description': self._html_search_meta(['description', 'og:description'], webpage), + 'url': smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % (account_id, video_id), {'geo_countries': ['JP']}), + 'ie_key': 'BrightcoveNew', + }