X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/fe979149c83b5a935f7d28baf75848a9137316fd..6245a4128588146e39f66b371624ee39bdbb7720:/youtube_dl/extractor/yahoo.py?ds=sidebyside diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index 927a964..238d9ce 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -1,357 +1,315 @@ # coding: utf-8 from __future__ import unicode_literals +import hashlib import itertools -import json import re from .common import InfoExtractor, SearchInfoExtractor from ..compat import ( + compat_str, compat_urllib_parse, - compat_urllib_parse_urlencode, - compat_urlparse, ) from ..utils import ( clean_html, - unescapeHTML, - ExtractorError, int_or_none, mimetype2ext, + parse_iso8601, + smuggle_url, + try_get, + url_or_none, ) -from .nbc import NBCSportsVPlayerIE +from .brightcove import BrightcoveNewIE class YahooIE(InfoExtractor): IE_DESC = 'Yahoo screen and movies' - _VALID_URL = r'(?P(?Phttps?://(?:[a-zA-Z]{2}\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?P.+)?-(?P[0-9]+)(?:-[a-z]+)?(?:\.html)?)' - _TESTS = [ - { - 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html', - 'info_dict': { - 'id': '2d25e626-2378-391f-ada0-ddaf1417e588', - 'ext': 'mp4', - 'title': 'Julian Smith & Travis Legg Watch Julian Smith', - 'description': 'Julian and Travis watch Julian Smith', - 'duration': 6863, - }, + _VALID_URL = r'(?Phttps?://(?:(?P[a-zA-Z]{2}(?:-[a-zA-Z]{2})?|malaysia)\.)?(?:[\da-zA-Z_-]+\.)?yahoo\.com/(?:[^/]+/)*(?P[^?&#]*-[0-9]+(?:-[a-z]+)?)\.html)' + _TESTS = [{ + 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html', + 'info_dict': { + 'id': '2d25e626-2378-391f-ada0-ddaf1417e588', + 'ext': 'mp4', + 'title': 'Julian Smith & Travis Legg Watch Julian Smith', + 'description': 'Julian and Travis watch Julian Smith', + 'duration': 6863, + 'timestamp': 1369812016, + 'upload_date': '20130529', }, - { - 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html', - 'md5': 'c3466d2b6d5dd6b9f41ba9ed04c24b23', - 'info_dict': { - 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9', - 'ext': 'mp4', - 'title': 'Codefellas - The Cougar Lies with Spanish Moss', - 'description': 'md5:66b627ab0a282b26352136ca96ce73c1', - 'duration': 151, - }, + }, { + 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed', + 'md5': '7993e572fac98e044588d0b5260f4352', + 'info_dict': { + 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb', + 'ext': 'mp4', + 'title': "Yahoo Saves 'Community'", + 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053', + 'duration': 170, + 'timestamp': 1406838636, + 'upload_date': '20140731', }, - { - 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed', - 'md5': '75ffabdb87c16d4ffe8c036dc4d1c136', - 'info_dict': { - 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb', - 'ext': 'mp4', - 'title': "Yahoo Saves 'Community'", - 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053', - 'duration': 170, - } + }, { + 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html', + 'md5': '71298482f7c64cbb7fa064e4553ff1c1', + 'info_dict': { + 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58', + 'ext': 'webm', + 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder', + 'description': 'md5:f66c890e1490f4910a9953c941dee944', + 'duration': 97, + 'timestamp': 1414489862, + 'upload_date': '20141028', + } + }, { + 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html', + 'md5': '88e209b417f173d86186bef6e4d1f160', + 'info_dict': { + 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521', + 'ext': 'mp4', + 'title': 'China Moses Is Crazy About the Blues', + 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0', + 'duration': 128, + 'timestamp': 1385722202, + 'upload_date': '20131129', + } + }, { + 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html', + 'md5': '2a9752f74cb898af5d1083ea9f661b58', + 'info_dict': { + 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', + 'ext': 'mp4', + 'title': '\'True Story\' Trailer', + 'description': 'True Story', + 'duration': 150, + 'timestamp': 1418919206, + 'upload_date': '20141218', }, - { - 'url': 'https://tw.news.yahoo.com/%E6%95%A2%E5%95%8F%E5%B8%82%E9%95%B7%20%E9%BB%83%E7%A7%80%E9%9C%9C%E6%89%B9%E8%B3%B4%E6%B8%85%E5%BE%B7%20%E9%9D%9E%E5%B8%B8%E9%AB%98%E5%82%B2-034024051.html', - 'md5': '9035d38f88b1782682a3e89f985be5bb', - 'info_dict': { - 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f', - 'ext': 'mp4', - 'title': '敢問市長／黃秀霜批賴清德「非常高傲」', - 'description': '直言台南沒捷運 交通居五都之末', - 'duration': 396, - }, + }, { + 'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html', + 'only_matching': True, + }, { + 'note': 'NBC Sports embeds', + 'url': 'http://sports.yahoo.com/blogs/ncaab-the-dagger/tyler-kalinoski-s-buzzer-beater-caps-davidson-s-comeback-win-185609842.html?guid=nbc_cbk_davidsonbuzzerbeater_150313', + 'info_dict': { + 'id': '9CsDKds0kvHI', + 'ext': 'flv', + 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d', + 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson', + 'upload_date': '20150313', + 'uploader': 'NBCU-SPORTS', + 'timestamp': 1426270238, }, - { - 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html', - 'md5': '0b51660361f0e27c9789e7037ef76f4b', - 'info_dict': { - 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58', - 'ext': 'mp4', - 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder', - 'description': 'md5:f66c890e1490f4910a9953c941dee944', - 'duration': 97, - } + }, { + 'url': 'https://tw.news.yahoo.com/-100120367.html', + 'only_matching': True, + }, { + # Query result is embedded in webpage, but explicit request to video API fails with geo restriction + 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html', + 'md5': '4fbafb9c9b6f07aa8f870629f6671b35', + 'info_dict': { + 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504', + 'ext': 'mp4', + 'title': 'Communitary - Community Episode 1: Ladders', + 'description': 'md5:8fc39608213295748e1e289807838c97', + 'duration': 1646, + 'timestamp': 1440436550, + 'upload_date': '20150824', + 'series': 'Communitary', + 'season_number': 6, + 'episode_number': 1, }, - { - 'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html', - 'md5': '57e06440778b1828a6079d2f744212c4', - 'info_dict': { - 'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73', - 'ext': 'mp4', - 'title': 'Program that makes hockey more affordable not offered in Manitoba', - 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4', - 'duration': 121, - }, - 'skip': 'Video gone', - }, { - 'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html', - 'info_dict': { - 'id': '154609075', - }, - 'playlist': [{ - 'md5': 'f8e336c6b66f503282e5f719641d6565', - 'info_dict': { - 'id': 'e624c4bc-3389-34de-9dfc-025f74943409', - 'ext': 'mp4', - 'title': '\'The Interview\' TV Spot: War', - 'description': 'The Interview', - 'duration': 30, - }, - }, { - 'md5': '958bcb90b4d6df71c56312137ee1cd5a', - 'info_dict': { - 'id': '1fc8ada0-718e-3abe-a450-bf31f246d1a9', - 'ext': 'mp4', - 'title': '\'The Interview\' TV Spot: Guys', - 'description': 'The Interview', - 'duration': 30, - }, - }], - }, { - 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html', - 'md5': '88e209b417f173d86186bef6e4d1f160', - 'info_dict': { - 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521', - 'ext': 'mp4', - 'title': 'China Moses Is Crazy About the Blues', - 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0', - 'duration': 128, - } - }, { - 'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html', - 'md5': 'd9a083ccf1379127bf25699d67e4791b', - 'info_dict': { - 'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c', - 'ext': 'mp4', - 'title': 'Connect the Dots: Dark Side of Virgo', - 'description': 'md5:1428185051cfd1949807ad4ff6d3686a', - 'duration': 201, - }, - 'skip': 'Domain name in.lifestyle.yahoo.com gone', - }, { - 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html', - 'md5': 'b17ac378b1134fa44370fb27db09a744', - 'info_dict': { - 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', - 'ext': 'mp4', - 'title': '\'True Story\' Trailer', - 'description': 'True Story', - 'duration': 150, - }, - }, { - 'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html', - 'only_matching': True, - }, { - 'note': 'NBC Sports embeds', - 'url': 'http://sports.yahoo.com/blogs/ncaab-the-dagger/tyler-kalinoski-s-buzzer-beater-caps-davidson-s-comeback-win-185609842.html?guid=nbc_cbk_davidsonbuzzerbeater_150313', - 'info_dict': { - 'id': '9CsDKds0kvHI', - 'ext': 'flv', - 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d', - 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson', - 'upload_date': '20150313', - 'uploader': 'NBCU-SPORTS', - 'timestamp': 1426270238, - } - }, { - 'url': 'https://tw.news.yahoo.com/-100120367.html', - 'only_matching': True, - }, { - # Query result is embedded in webpage, but explicit request to video API fails with geo restriction - 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html', - 'md5': '1ddbf7c850777548438e5c4f147c7b8c', + }, { + # ytwnews://cavideo/ + 'url': 'https://tw.video.yahoo.com/movie-tw/單車天使-中文版預-092316541.html', + 'info_dict': { + 'id': 'ba133ff2-0793-3510-b636-59dfe9ff6cff', + 'ext': 'mp4', + 'title': '單車天使 - 中文版預', + 'description': '中文版預', + 'timestamp': 1476696196, + 'upload_date': '20161017', + }, + 'params': { + 'skip_download': True, + }, + }, { + # Contains both a Yahoo hosted video and multiple Youtube embeds + 'url': 'https://www.yahoo.com/entertainment/gwen-stefani-reveals-the-pop-hit-she-passed-on-assigns-it-to-her-voice-contestant-instead-033045672.html', + 'info_dict': { + 'id': '46c5d95a-528f-3d03-b732-732fcadd51de', + 'title': 'Gwen Stefani reveals the pop hit she passed on, assigns it to her \'Voice\' contestant instead', + 'description': 'Gwen decided not to record this hit herself, but she decided it was the perfect fit for Kyndall Inskeep.', + }, + 'playlist': [{ 'info_dict': { - 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504', + 'id': '966d4262-4fd1-3aaa-b45b-049ca6e38ba6', 'ext': 'mp4', - 'title': 'Communitary - Community Episode 1: Ladders', - 'description': 'md5:8fc39608213295748e1e289807838c97', - 'duration': 1646, + 'title': 'Gwen Stefani reveals she turned down one of Sia\'s best songs', + 'description': 'On "The Voice" Tuesday, Gwen Stefani told Taylor Swift which Sia hit was almost hers.', + 'timestamp': 1572406500, + 'upload_date': '20191030', }, }, { - # it uses an alias to get the video_id - 'url': 'https://www.yahoo.com/movies/the-stars-of-daddys-home-have-very-different-212843197.html', 'info_dict': { - 'id': '40eda9c8-8e5f-3552-8745-830f67d0c737', + 'id': '352CFDOQrKg', 'ext': 'mp4', - 'title': 'Will Ferrell & Mark Wahlberg Are Pro-Spanking', - 'description': 'While they play feuding fathers in \'Daddy\'s Home,\' star Will Ferrell & Mark Wahlberg share their true feelings on parenthood.', + 'title': 'Kyndal Inskeep "Performs the Hell Out of" Sia\'s "Elastic Heart" - The Voice Knockouts 2019', + 'description': 'md5:35b61e94c2ae214bc965ff4245f80d11', + 'uploader': 'The Voice', + 'uploader_id': 'NBCTheVoice', + 'upload_date': '20191029', }, + }], + 'params': { + 'playlistend': 2, }, - { - # config['models']['applet_model']['data']['sapi'] has no query - 'url': 'https://www.yahoo.com/music/livenation/event/galactic-2016', - 'md5': 'dac0c72d502bc5facda80c9e6d5c98db', - 'info_dict': { - 'id': 'a6015640-e9e5-3efb-bb60-05589a183919', - 'ext': 'mp4', - 'description': 'Galactic', - 'title': 'Dolla Diva (feat. Maggie Koerner)', - }, - }, - ] + 'expected_warnings': ['HTTP Error 404'], + }, { + 'url': 'https://malaysia.news.yahoo.com/video/bystanders-help-ontario-policeman-bust-190932818.html', + 'only_matching': True, + }, { + 'url': 'https://es-us.noticias.yahoo.com/es-la-puerta-irrompible-que-110539379.html', + 'only_matching': True, + }, { + 'url': 'https://www.yahoo.com/entertainment/v/longtime-cbs-news-60-minutes-032036500-cbs.html', + 'only_matching': True, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('display_id') or self._match_id(url) - page_id = mobj.group('id') - url = mobj.group('url') - host = mobj.group('host') - webpage, urlh = self._download_webpage_handle(url, display_id) - if 'err=404' in urlh.geturl(): - raise ExtractorError('Video gone', expected=True) - - # Look for iframed media first - entries = [] - iframe_urls = re.findall(r']+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage) - for idx, iframe_url in enumerate(iframe_urls): - iframepage = self._download_webpage( - host + iframe_url, display_id, - note='Downloading iframe webpage for video #%d' % idx) - items_json = self._search_regex( - r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None) - if items_json: - items = json.loads(items_json) - video_id = items[0]['id'] - entries.append(self._get_info(video_id, display_id, webpage)) - if entries: - return self.playlist_result(entries, page_id) - - # Look for NBCSports iframes - nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage) - if nbc_sports_url: - return self.url_result(nbc_sports_url, 'NBCSportsVPlayer') - - # Query result is often embedded in webpage as JSON. Sometimes explicit requests - # to video API results in a failure with geo restriction reason therefore using - # embedded query result when present sounds reasonable. - config_json = self._search_regex( - r'window\.Af\.bootstrap\[[^\]]+\]\s*=\s*({.*?"applet_type"\s*:\s*"td-applet-videoplayer".*?});(?:|$)', - webpage, 'videoplayer applet', default=None) - if config_json: - config = self._parse_json(config_json, display_id, fatal=False) - if config: - sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi') - if sapi and 'query' in sapi: - return self._extract_info(display_id, sapi, webpage) - - items_json = self._search_regex( - r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE, - default=None) - if items_json is None: - alias = self._search_regex( - r'"aliases":{"video":"(.*?)"', webpage, 'alias', default=None) - if alias is not None: - alias_info = self._download_json( - 'https://www.yahoo.com/_td/api/resource/VideoService.videos;video_aliases=["%s"]' % alias, - display_id, 'Downloading alias info') - video_id = alias_info[0]['id'] - else: - CONTENT_ID_REGEXES = [ - r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"', - r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"', - r'"first_videoid"\s*:\s*"([^"]+)"', - r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id), - r']data-uuid=["\']([^"\']+)', - r'yahoo://article/view\?.*\buuid=([^&"\']+)', - ] - video_id = self._search_regex( - CONTENT_ID_REGEXES, webpage, 'content ID') + url, country, display_id = re.match(self._VALID_URL, url).groups() + if not country: + country = 'us' else: - items = json.loads(items_json) - info = items['mediaItems']['query']['results']['mediaObj'][0] - # The 'meta' field is not always in the video webpage, we request it - # from another page - video_id = info['id'] - return self._get_info(video_id, display_id, webpage) - - def _extract_info(self, display_id, query, webpage): - info = query['query']['results']['mediaObj'][0] - meta = info.get('meta') - video_id = info.get('id') - - if not meta: - msg = info['status'].get('msg') - if msg: - raise ExtractorError( - '%s returned error: %s' % (self.IE_NAME, msg), expected=True) - raise ExtractorError('Unable to extract media object meta') + country = country.split('-')[0] + api_base = 'https://%s.yahoo.com/_td/api/resource/' % country + + for i, uuid in enumerate(['url=' + url, 'ymedia-alias=' + display_id]): + content = self._download_json( + api_base + 'content;getDetailView=true;uuids=["%s"]' % uuid, + display_id, 'Downloading content JSON metadata', fatal=i == 1) + if content: + item = content['items'][0] + break + + if item.get('type') != 'video': + entries = [] + + cover = item.get('cover') or {} + if cover.get('type') == 'yvideo': + cover_url = cover.get('url') + if cover_url: + entries.append(self.url_result( + cover_url, 'Yahoo', cover.get('uuid'))) + + for e in item.get('body', []): + if e.get('type') == 'videoIframe': + iframe_url = e.get('url') + if not iframe_url: + continue + entries.append(self.url_result(iframe_url)) + + return self.playlist_result( + entries, item.get('uuid'), + item.get('title'), item.get('summary')) + video_id = item['uuid'] + video = self._download_json( + api_base + 'VideoService.videos;view=full;video_ids=["%s"]' % video_id, + video_id, 'Downloading video JSON metadata')[0] + title = video['title'] + + if country == 'malaysia': + country = 'my' + + is_live = video.get('live_state') == 'live' + fmts = ('m3u8',) if is_live else ('webm', 'mp4') + + urls = [] formats = [] - for s in info['streams']: - format_info = { - 'width': int_or_none(s.get('width')), - 'height': int_or_none(s.get('height')), - 'tbr': int_or_none(s.get('bitrate')), - } - - host = s['host'] - path = s['path'] - if host.startswith('rtmp'): - format_info.update({ - 'url': host, - 'play_path': path, - 'ext': 'flv', + subtitles = {} + for fmt in fmts: + media_obj = self._download_json( + 'https://video-api.yql.yahoo.com/v1/video/sapi/streams/' + video_id, + video_id, 'Downloading %s JSON metadata' % fmt, + headers=self.geo_verification_headers(), query={ + 'format': fmt, + 'region': country.upper(), + })['query']['results']['mediaObj'][0] + msg = media_obj.get('status', {}).get('msg') + + for s in media_obj.get('streams', []): + host = s.get('host') + path = s.get('path') + if not host or not path: + continue + s_url = host + path + if s.get('format') == 'm3u8': + formats.extend(self._extract_m3u8_formats( + s_url, video_id, 'mp4', m3u8_id='hls', fatal=False)) + continue + tbr = int_or_none(s.get('bitrate')) + formats.append({ + 'url': s_url, + 'format_id': fmt + ('-%d' % tbr if tbr else ''), + 'width': int_or_none(s.get('width')), + 'height': int_or_none(s.get('height')), + 'tbr': tbr, + 'fps': int_or_none(s.get('framerate')), }) - else: - if s.get('format') == 'm3u8_playlist': - format_info['protocol'] = 'm3u8_native' - format_info['ext'] = 'mp4' - format_url = compat_urlparse.urljoin(host, path) - format_info['url'] = format_url - formats.append(format_info) + + for cc in media_obj.get('closedcaptions', []): + cc_url = cc.get('url') + if not cc_url or cc_url in urls: + continue + urls.append(cc_url) + subtitles.setdefault(cc.get('lang') or 'en-US', []).append({ + 'url': cc_url, + 'ext': mimetype2ext(cc.get('content_type')), + }) + + streaming_url = video.get('streaming_url') + if streaming_url and not is_live: + formats.extend(self._extract_m3u8_formats( + streaming_url, video_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False)) + + if not formats and msg == 'geo restricted': + self.raise_geo_restricted() self._sort_formats(formats) - closed_captions = self._html_search_regex( - r'"closedcaptions":(\[[^\]]+\])', webpage, 'closed captions', - default='[]') + thumbnails = [] + for thumb in video.get('thumbnails', []): + thumb_url = thumb.get('url') + if not thumb_url: + continue + thumbnails.append({ + 'id': thumb.get('tag'), + 'url': thumb.get('url'), + 'width': int_or_none(thumb.get('width')), + 'height': int_or_none(thumb.get('height')), + }) - cc_json = self._parse_json(closed_captions, video_id, fatal=False) - subtitles = {} - if cc_json: - for closed_caption in cc_json: - lang = closed_caption['lang'] - if lang not in subtitles: - subtitles[lang] = [] - subtitles[lang].append({ - 'url': closed_caption['url'], - 'ext': mimetype2ext(closed_caption['content_type']), - }) + series_info = video.get('series_info') or {} return { 'id': video_id, - 'display_id': display_id, - 'title': unescapeHTML(meta['title']), + 'title': self._live_title(title) if is_live else title, 'formats': formats, - 'description': clean_html(meta['description']), - 'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage), - 'duration': int_or_none(meta.get('duration')), + 'display_id': display_id, + 'thumbnails': thumbnails, + 'description': clean_html(video.get('description')), + 'timestamp': parse_iso8601(video.get('publish_time')), 'subtitles': subtitles, + 'duration': int_or_none(video.get('duration')), + 'view_count': int_or_none(video.get('view_count')), + 'is_live': is_live, + 'series': video.get('show_name'), + 'season_number': int_or_none(series_info.get('season_number')), + 'episode_number': int_or_none(series_info.get('episode_number')), } - def _get_info(self, video_id, display_id, webpage): - region = self._search_regex( - r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"', - webpage, 'region', fatal=False, default='US') - data = compat_urllib_parse_urlencode({ - 'protocol': 'http', - 'region': region.upper(), - }) - query_url = ( - 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' - '{id}?{data}'.format(id=video_id, data=data)) - query_result = self._download_json( - query_url, display_id, 'Downloading video info') - return self._extract_info(display_id, query_result, webpage) - class YahooSearchIE(SearchInfoExtractor): IE_DESC = 'Yahoo screen search' @@ -383,3 +341,209 @@ class YahooSearchIE(SearchInfoExtractor): 'id': query, 'entries': entries, } + + +class YahooGyaOPlayerIE(InfoExtractor): + IE_NAME = 'yahoo:gyao:player' + _VALID_URL = r'https?://(?:gyao\.yahoo\.co\.jp/(?:player|episode/[^/]+)|streaming\.yahoo\.co\.jp/c/y)/(?P\d+/v\d+/v\d+|[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})' + _TESTS = [{ + 'url': 'https://gyao.yahoo.co.jp/player/00998/v00818/v0000000000000008564/', + 'info_dict': { + 'id': '5993125228001', + 'ext': 'mp4', + 'title': 'フューリー　【字幕版】', + 'description': 'md5:21e691c798a15330eda4db17a8fe45a5', + 'uploader_id': '4235717419001', + 'upload_date': '20190124', + 'timestamp': 1548294365, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + 'url': 'https://streaming.yahoo.co.jp/c/y/01034/v00133/v0000000000000000706/', + 'only_matching': True, + }, { + 'url': 'https://gyao.yahoo.co.jp/episode/%E3%81%8D%E3%81%AE%E3%81%86%E4%BD%95%E9%A3%9F%E3%81%B9%E3%81%9F%EF%BC%9F%20%E7%AC%AC2%E8%A9%B1%202019%2F4%2F12%E6%94%BE%E9%80%81%E5%88%86/5cb02352-b725-409e-9f8d-88f947a9f682', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url).replace('/', ':') + video = self._download_json( + 'https://gyao.yahoo.co.jp/dam/v1/videos/' + video_id, + video_id, query={ + 'fields': 'longDescription,title,videoId', + }, headers={ + 'X-User-Agent': 'Unknown Pc GYAO!/2.0.0 Web', + }) + return { + '_type': 'url_transparent', + 'id': video_id, + 'title': video['title'], + 'url': smuggle_url( + 'http://players.brightcove.net/4235717419001/SyG5P0gjb_default/index.html?videoId=' + video['videoId'], + {'geo_countries': ['JP']}), + 'description': video.get('longDescription'), + 'ie_key': BrightcoveNewIE.ie_key(), + } + + +class YahooGyaOIE(InfoExtractor): + IE_NAME = 'yahoo:gyao' + _VALID_URL = r'https?://(?:gyao\.yahoo\.co\.jp/(?:p|title/[^/]+)|streaming\.yahoo\.co\.jp/p/y)/(?P\d+/v\d+|[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})' + _TESTS = [{ + 'url': 'https://gyao.yahoo.co.jp/p/00449/v03102/', + 'info_dict': { + 'id': '00449:v03102', + }, + 'playlist_count': 2, + }, { + 'url': 'https://streaming.yahoo.co.jp/p/y/01034/v00133/', + 'only_matching': True, + }, { + 'url': 'https://gyao.yahoo.co.jp/title/%E3%81%97%E3%82%83%E3%81%B9%E3%81%8F%E3%82%8A007/5b025a49-b2e5-4dc7-945c-09c6634afacf', + 'only_matching': True, + }] + + def _real_extract(self, url): + program_id = self._match_id(url).replace('/', ':') + videos = self._download_json( + 'https://gyao.yahoo.co.jp/api/programs/%s/videos' % program_id, program_id)['videos'] + entries = [] + for video in videos: + video_id = video.get('id') + if not video_id: + continue + entries.append(self.url_result( + 'https://gyao.yahoo.co.jp/player/%s/' % video_id.replace(':', '/'), + YahooGyaOPlayerIE.ie_key(), video_id)) + return self.playlist_result(entries, program_id) + + +class YahooJapanNewsIE(InfoExtractor): + IE_NAME = 'yahoo:japannews' + IE_DESC = 'Yahoo! Japan News' + _VALID_URL = r'https?://(?P(?:news|headlines)\.yahoo\.co\.jp)[^\d]*(?P\d[\d-]*\d)?' + _GEO_COUNTRIES = ['JP'] + _TESTS = [{ + 'url': 'https://headlines.yahoo.co.jp/videonews/ann?a=20190716-00000071-ann-int', + 'info_dict': { + 'id': '1736242', + 'ext': 'mp4', + 'title': 'ムン大統領が対日批判を強化“現金化”効果は？（テレビ朝日系（ANN）） - Yahoo!ニュース', + 'description': '韓国の元徴用工らを巡る裁判の原告が弁護士が差し押さえた三菱重工業の資産を売却して - Yahoo!ニュース(テレビ朝日系（ANN）)', + 'thumbnail': r're:^https?://.*\.[a-zA-Z\d]{3,4}$', + }, + 'params': { + 'skip_download': True, + }, + }, { + # geo restricted + 'url': 'https://headlines.yahoo.co.jp/hl?a=20190721-00000001-oxv-l04', + 'only_matching': True, + }, { + 'url': 'https://headlines.yahoo.co.jp/videonews/', + 'only_matching': True, + }, { + 'url': 'https://news.yahoo.co.jp', + 'only_matching': True, + }, { + 'url': 'https://news.yahoo.co.jp/byline/hashimotojunji/20190628-00131977/', + 'only_matching': True, + }, { + 'url': 'https://news.yahoo.co.jp/feature/1356', + 'only_matching': True + }] + + def _extract_formats(self, json_data, content_id): + formats = [] + + video_data = try_get( + json_data, + lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'], + list) + for vid in video_data or []: + delivery = vid.get('delivery') + url = url_or_none(vid.get('Url')) + if not delivery or not url: + continue + elif delivery == 'hls': + formats.extend( + self._extract_m3u8_formats( + url, content_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + else: + formats.append({ + 'url': url, + 'format_id': 'http-%s' % compat_str(vid.get('bitrate', '')), + 'height': int_or_none(vid.get('height')), + 'width': int_or_none(vid.get('width')), + 'tbr': int_or_none(vid.get('bitrate')), + }) + self._remove_duplicate_formats(formats) + self._sort_formats(formats) + + return formats + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + host = mobj.group('host') + display_id = mobj.group('id') or host + + webpage = self._download_webpage(url, display_id) + + title = self._html_search_meta( + ['og:title', 'twitter:title'], webpage, 'title', default=None + ) or self._html_search_regex('([^<]+)', webpage, 'title') + + if display_id == host: + # Headline page (w/ multiple BC playlists) ('news.yahoo.co.jp', 'headlines.yahoo.co.jp/videonews/', ...) + stream_plists = re.findall(r'plist=(\d+)', webpage) or re.findall(r'plist["\']:\s*["\']([^"\']+)', webpage) + entries = [ + self.url_result( + smuggle_url( + 'http://players.brightcove.net/5690807595001/HyZNerRl7_default/index.html?playlistId=%s' % plist_id, + {'geo_countries': ['JP']}), + ie='BrightcoveNew', video_id=plist_id) + for plist_id in stream_plists] + return self.playlist_result(entries, playlist_title=title) + + # Article page + description = self._html_search_meta( + ['og:description', 'description', 'twitter:description'], + webpage, 'description', default=None) + thumbnail = self._og_search_thumbnail( + webpage, default=None) or self._html_search_meta( + 'twitter:image', webpage, 'thumbnail', default=None) + space_id = self._search_regex([ + r']+class=["\']yvpub-player["\'][^>]+spaceid=([^&"\']+)', + r'YAHOO\.JP\.srch\.\w+link\.onLoad[^;]+spaceID["\' ]*:["\' ]+([^"\']+)', + r'