X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/ced7488f6d3a519b2c1b1cbd31048743fb8285bd..4d055bc6eb5dbe9a42216ec3e659d8ff1f2a1256:/youtube_dl/extractor/yahoo.py diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index b0679df..e5ebdd1 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -1,31 +1,40 @@ # 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, + determine_ext, ExtractorError, + extract_attributes, int_or_none, mimetype2ext, + smuggle_url, + try_get, + unescapeHTML, + url_or_none, ) -from .brightcove import BrightcoveNewIE +from .brightcove import ( + BrightcoveLegacyIE, + BrightcoveNewIE, +) from .nbc import NBCSportsVPlayerIE 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)?)' + _VALID_URL = r'(?Phttps?://(?:(?P[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', @@ -39,7 +48,7 @@ class YahooIE(InfoExtractor): }, { 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html', - 'md5': 'c3466d2b6d5dd6b9f41ba9ed04c24b23', + 'md5': '251af144a19ebc4a033e8ba91ac726bb', 'info_dict': { 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9', 'ext': 'mp4', @@ -47,10 +56,11 @@ class YahooIE(InfoExtractor): 'description': 'md5:66b627ab0a282b26352136ca96ce73c1', 'duration': 151, }, + 'skip': 'HTTP Error 404', }, { 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed', - 'md5': '75ffabdb87c16d4ffe8c036dc4d1c136', + 'md5': '7993e572fac98e044588d0b5260f4352', 'info_dict': { 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb', 'ext': 'mp4', @@ -61,7 +71,7 @@ class YahooIE(InfoExtractor): }, { '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', + 'md5': '45c024bad51e63e9b6f6fad7a43a8c23', 'info_dict': { 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f', 'ext': 'mp4', @@ -72,10 +82,10 @@ class YahooIE(InfoExtractor): }, { 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html', - 'md5': '0b51660361f0e27c9789e7037ef76f4b', + 'md5': '71298482f7c64cbb7fa064e4553ff1c1', 'info_dict': { 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58', - 'ext': 'mp4', + 'ext': 'webm', 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder', 'description': 'md5:f66c890e1490f4910a9953c941dee944', 'duration': 97, @@ -98,7 +108,7 @@ class YahooIE(InfoExtractor): 'id': '154609075', }, 'playlist': [{ - 'md5': 'f8e336c6b66f503282e5f719641d6565', + 'md5': '000887d0dc609bc3a47c974151a40fb8', 'info_dict': { 'id': 'e624c4bc-3389-34de-9dfc-025f74943409', 'ext': 'mp4', @@ -107,7 +117,7 @@ class YahooIE(InfoExtractor): 'duration': 30, }, }, { - 'md5': '958bcb90b4d6df71c56312137ee1cd5a', + 'md5': '81bc74faf10750fe36e4542f9a184c66', 'info_dict': { 'id': '1fc8ada0-718e-3abe-a450-bf31f246d1a9', 'ext': 'mp4', @@ -139,7 +149,7 @@ class YahooIE(InfoExtractor): 'skip': 'Domain name in.lifestyle.yahoo.com gone', }, { 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html', - 'md5': 'b17ac378b1134fa44370fb27db09a744', + 'md5': '989396ae73d20c6f057746fb226aa215', 'info_dict': { 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', 'ext': 'mp4', @@ -168,7 +178,7 @@ class YahooIE(InfoExtractor): }, { # 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', + 'md5': '4fbafb9c9b6f07aa8f870629f6671b35', 'info_dict': { 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504', 'ext': 'mp4', @@ -196,14 +206,61 @@ class YahooIE(InfoExtractor): 'description': 'Galactic', 'title': 'Dolla Diva (feat. Maggie Koerner)', }, + 'skip': 'redirect to https://www.yahoo.com/music', + }, + { + # yahoo://article/ + 'url': 'https://www.yahoo.com/movies/video/true-story-trailer-173000497.html', + 'info_dict': { + 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', + 'ext': 'mp4', + 'title': "'True Story' Trailer", + 'description': 'True Story', + }, + 'params': { + 'skip_download': True, + }, + }, + { + # 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', + }, + 'params': { + 'skip_download': True, + }, + }, + { + # custom brightcove + 'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37083565/clown-entertainers-say-it-is-hurting-their-business/', + 'info_dict': { + 'id': '5575377707001', + 'ext': 'mp4', + 'title': "Clown entertainers say 'It' is hurting their business", + 'description': 'Stephen King s horror film has much to answer for. Jelby and Mr Loopy the Clowns join us.', + 'timestamp': 1505341164, + 'upload_date': '20170913', + 'uploader_id': '2376984109001', + }, + 'params': { + 'skip_download': True, + }, }, + { + # custom brightcove, geo-restricted to Australia, bypassable + 'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37263964/sunrise-episode-wed-27-sep/', + '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') + display_id = mobj.group('display_id') or page_id host = mobj.group('host') webpage, urlh = self._download_webpage_handle(url, display_id) if 'err=404' in urlh.geturl(): @@ -213,15 +270,7 @@ class YahooIE(InfoExtractor): 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)) + entries.append(self.url_result(host + iframe_url, 'Yahoo')) if entries: return self.playlist_result(entries, page_id) @@ -230,10 +279,36 @@ class YahooIE(InfoExtractor): if nbc_sports_url: return self.url_result(nbc_sports_url, NBCSportsVPlayerIE.ie_key()) + # Look for Brightcove Legacy Studio embeds + bc_url = BrightcoveLegacyIE._extract_brightcove_url(webpage) + if bc_url: + return self.url_result(bc_url, BrightcoveLegacyIE.ie_key()) + + def brightcove_url_result(bc_url): + return self.url_result( + smuggle_url(bc_url, {'geo_countries': [mobj.group('country')]}), + BrightcoveNewIE.ie_key()) + # Look for Brightcove New Studio embeds - bc_url = BrightcoveNewIE._extract_url(webpage) + bc_url = BrightcoveNewIE._extract_url(self, webpage) if bc_url: - return self.url_result(bc_url, BrightcoveNewIE.ie_key()) + return brightcove_url_result(bc_url) + + brightcove_iframe = self._search_regex( + r'(]+data-video-id=["\']\d+[^>]+>)', webpage, + 'brightcove iframe', default=None) + if brightcove_iframe: + attr = extract_attributes(brightcove_iframe) + src = attr.get('src') + if src: + parsed_src = compat_urlparse.urlparse(src) + qs = compat_urlparse.parse_qs(parsed_src.query) + account_id = qs.get('accountId', ['2376984109001'])[0] + brightcove_id = attr.get('data-video-id') or qs.get('videoId', [None])[0] + if account_id and brightcove_id: + return brightcove_url_result( + 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s' + % (account_id, brightcove_id)) # 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 @@ -246,7 +321,9 @@ class YahooIE(InfoExtractor): 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) + info = self._extract_info(display_id, sapi, webpage) + self._sort_formats(info['formats']) + return info items_json = self._search_regex( r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE, @@ -266,7 +343,8 @@ class YahooIE(InfoExtractor): r'"first_videoid"\s*:\s*"([^"]+)"', r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id), r']data-uuid=["\']([^"\']+)', - r'yahoo://article/view\?.*\buuid=([^&"\']+)', + r']+yahoo://article/view\?.*\buuid=([^&"\']+)', + r']+["\']ytwnews://cavideo/(?:[^/]+/)+([\da-fA-F-]+)[&"\']', ] video_id = self._search_regex( CONTENT_ID_REGEXES, webpage, 'content ID') @@ -292,15 +370,17 @@ class YahooIE(InfoExtractor): formats = [] for s in info['streams']: + tbr = int_or_none(s.get('bitrate')) format_info = { 'width': int_or_none(s.get('width')), 'height': int_or_none(s.get('height')), - 'tbr': int_or_none(s.get('bitrate')), + 'tbr': tbr, } host = s['host'] path = s['path'] if host.startswith('rtmp'): + fmt = 'rtmp' format_info.update({ 'url': host, 'play_path': path, @@ -308,14 +388,18 @@ class YahooIE(InfoExtractor): }) else: if s.get('format') == 'm3u8_playlist': - format_info['protocol'] = 'm3u8_native' - format_info['ext'] = 'mp4' + fmt = 'hls' + format_info.update({ + 'protocol': 'm3u8_native', + 'ext': 'mp4', + }) + else: + fmt = format_info['ext'] = determine_ext(path) format_url = compat_urlparse.urljoin(host, path) format_info['url'] = format_url + format_info['format_id'] = fmt + ('-%d' % tbr if tbr else '') formats.append(format_info) - self._sort_formats(formats) - closed_captions = self._html_search_regex( r'"closedcaptions":(\[[^\]]+\])', webpage, 'closed captions', default='[]') @@ -346,17 +430,25 @@ class YahooIE(InfoExtractor): 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) + webpage, 'region', fatal=False, default='US').upper() + formats = [] + info = {} + for fmt in ('webm', 'mp4'): + query_result = self._download_json( + 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' + video_id, + display_id, 'Downloading %s video info' % fmt, query={ + 'protocol': 'http', + 'region': region, + 'format': fmt, + }) + info = self._extract_info(display_id, query_result, webpage) + formats.extend(info['formats']) + formats.extend(self._extract_m3u8_formats( + 'http://video.media.yql.yahoo.com/v1/hls/%s?region=%s' % (video_id, region), + video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) + self._sort_formats(formats) + info['formats'] = formats + return info class YahooSearchIE(SearchInfoExtractor): @@ -389,3 +481,209 @@ 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/default_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'