Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/yahoo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import hashlib
   5 import itertools
   6 import json
   7 import re
   8
   9 from .common import InfoExtractor, SearchInfoExtractor
  10 from ..compat import (
  11     compat_str,
  12     compat_urllib_parse,
  13     compat_urlparse,
  14 )
  15 from ..utils import (
  16     clean_html,
  17     determine_ext,
  18     ExtractorError,
  19     extract_attributes,
  20     int_or_none,
  21     mimetype2ext,
  22     smuggle_url,
  23     try_get,
  24     unescapeHTML,
  25     url_or_none,
  26 )
  27
  28 from .brightcove import (
  29     BrightcoveLegacyIE,
  30     BrightcoveNewIE,
  31 )
  32 from .nbc import NBCSportsVPlayerIE
  33
  34
  35 class YahooIE(InfoExtractor):
  36     IE_DESC = 'Yahoo screen and movies'
  37     _VALID_URL = r'(?P<host>https?://(?:(?P<country>[a-zA-Z]{2})\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?:(?P<display_id>.+)?-)?(?P<id>[0-9]+)(?:-[a-z]+)?(?:\.html)?'
  38     _TESTS = [
  39         {
  40             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
  41             'info_dict': {
  42                 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
  43                 'ext': 'mp4',
  44                 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
  45                 'description': 'Julian and Travis watch Julian Smith',
  46                 'duration': 6863,
  47             },
  48         },
  49         {
  50             'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
  51             'md5': '251af144a19ebc4a033e8ba91ac726bb',
  52             'info_dict': {
  53                 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
  54                 'ext': 'mp4',
  55                 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
  56                 'description': 'md5:66b627ab0a282b26352136ca96ce73c1',
  57                 'duration': 151,
  58             },
  59             'skip': 'HTTP Error 404',
  60         },
  61         {
  62             'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
  63             'md5': '7993e572fac98e044588d0b5260f4352',
  64             'info_dict': {
  65                 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
  66                 'ext': 'mp4',
  67                 'title': "Yahoo Saves 'Community'",
  68                 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
  69                 'duration': 170,
  70             }
  71         },
  72         {
  73             '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',
  74             'md5': '45c024bad51e63e9b6f6fad7a43a8c23',
  75             'info_dict': {
  76                 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f',
  77                 'ext': 'mp4',
  78                 'title': '敢問市長／黃秀霜批賴清德「非常高傲」',
  79                 'description': '直言台南沒捷運 交通居五都之末',
  80                 'duration': 396,
  81             },
  82         },
  83         {
  84             'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
  85             'md5': '71298482f7c64cbb7fa064e4553ff1c1',
  86             'info_dict': {
  87                 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
  88                 'ext': 'webm',
  89                 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
  90                 'description': 'md5:f66c890e1490f4910a9953c941dee944',
  91                 'duration': 97,
  92             }
  93         },
  94         {
  95             'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html',
  96             'md5': '57e06440778b1828a6079d2f744212c4',
  97             'info_dict': {
  98                 'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73',
  99                 'ext': 'mp4',
 100                 'title': 'Program that makes hockey more affordable not offered in Manitoba',
 101                 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4',
 102                 'duration': 121,
 103             },
 104             'skip': 'Video gone',
 105         }, {
 106             'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html',
 107             'info_dict': {
 108                 'id': '154609075',
 109             },
 110             'playlist': [{
 111                 'md5': '000887d0dc609bc3a47c974151a40fb8',
 112                 'info_dict': {
 113                     'id': 'e624c4bc-3389-34de-9dfc-025f74943409',
 114                     'ext': 'mp4',
 115                     'title': '\'The Interview\' TV Spot: War',
 116                     'description': 'The Interview',
 117                     'duration': 30,
 118                 },
 119             }, {
 120                 'md5': '81bc74faf10750fe36e4542f9a184c66',
 121                 'info_dict': {
 122                     'id': '1fc8ada0-718e-3abe-a450-bf31f246d1a9',
 123                     'ext': 'mp4',
 124                     'title': '\'The Interview\' TV Spot: Guys',
 125                     'description': 'The Interview',
 126                     'duration': 30,
 127                 },
 128             }],
 129         }, {
 130             'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
 131             'md5': '88e209b417f173d86186bef6e4d1f160',
 132             'info_dict': {
 133                 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
 134                 'ext': 'mp4',
 135                 'title': 'China Moses Is Crazy About the Blues',
 136                 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
 137                 'duration': 128,
 138             }
 139         }, {
 140             'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html',
 141             'md5': 'd9a083ccf1379127bf25699d67e4791b',
 142             'info_dict': {
 143                 'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c',
 144                 'ext': 'mp4',
 145                 'title': 'Connect the Dots: Dark Side of Virgo',
 146                 'description': 'md5:1428185051cfd1949807ad4ff6d3686a',
 147                 'duration': 201,
 148             },
 149             'skip': 'Domain name in.lifestyle.yahoo.com gone',
 150         }, {
 151             'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
 152             'md5': '989396ae73d20c6f057746fb226aa215',
 153             'info_dict': {
 154                 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
 155                 'ext': 'mp4',
 156                 'title': '\'True Story\' Trailer',
 157                 'description': 'True Story',
 158                 'duration': 150,
 159             },
 160         }, {
 161             'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
 162             'only_matching': True,
 163         }, {
 164             'note': 'NBC Sports embeds',
 165             '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',
 166             'info_dict': {
 167                 'id': '9CsDKds0kvHI',
 168                 'ext': 'flv',
 169                 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
 170                 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
 171                 'upload_date': '20150313',
 172                 'uploader': 'NBCU-SPORTS',
 173                 'timestamp': 1426270238,
 174             }
 175         }, {
 176             'url': 'https://tw.news.yahoo.com/-100120367.html',
 177             'only_matching': True,
 178         }, {
 179             # Query result is embedded in webpage, but explicit request to video API fails with geo restriction
 180             'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html',
 181             'md5': '4fbafb9c9b6f07aa8f870629f6671b35',
 182             'info_dict': {
 183                 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504',
 184                 'ext': 'mp4',
 185                 'title': 'Communitary - Community Episode 1: Ladders',
 186                 'description': 'md5:8fc39608213295748e1e289807838c97',
 187                 'duration': 1646,
 188             },
 189         }, {
 190             # it uses an alias to get the video_id
 191             'url': 'https://www.yahoo.com/movies/the-stars-of-daddys-home-have-very-different-212843197.html',
 192             'info_dict': {
 193                 'id': '40eda9c8-8e5f-3552-8745-830f67d0c737',
 194                 'ext': 'mp4',
 195                 'title': 'Will Ferrell & Mark Wahlberg Are Pro-Spanking',
 196                 'description': 'While they play feuding fathers in \'Daddy\'s Home,\' star Will Ferrell & Mark Wahlberg share their true feelings on parenthood.',
 197             },
 198         },
 199         {
 200             # config['models']['applet_model']['data']['sapi'] has no query
 201             'url': 'https://www.yahoo.com/music/livenation/event/galactic-2016',
 202             'md5': 'dac0c72d502bc5facda80c9e6d5c98db',
 203             'info_dict': {
 204                 'id': 'a6015640-e9e5-3efb-bb60-05589a183919',
 205                 'ext': 'mp4',
 206                 'description': 'Galactic',
 207                 'title': 'Dolla Diva (feat. Maggie Koerner)',
 208             },
 209             'skip': 'redirect to https://www.yahoo.com/music',
 210         },
 211         {
 212             # yahoo://article/
 213             'url': 'https://www.yahoo.com/movies/video/true-story-trailer-173000497.html',
 214             'info_dict': {
 215                 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
 216                 'ext': 'mp4',
 217                 'title': "'True Story' Trailer",
 218                 'description': 'True Story',
 219             },
 220             'params': {
 221                 'skip_download': True,
 222             },
 223         },
 224         {
 225             # ytwnews://cavideo/
 226             'url': 'https://tw.video.yahoo.com/movie-tw/單車天使-中文版預-092316541.html',
 227             'info_dict': {
 228                 'id': 'ba133ff2-0793-3510-b636-59dfe9ff6cff',
 229                 'ext': 'mp4',
 230                 'title': '單車天使 - 中文版預',
 231                 'description': '中文版預',
 232             },
 233             'params': {
 234                 'skip_download': True,
 235             },
 236         },
 237         {
 238             # custom brightcove
 239             'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37083565/clown-entertainers-say-it-is-hurting-their-business/',
 240             'info_dict': {
 241                 'id': '5575377707001',
 242                 'ext': 'mp4',
 243                 'title': "Clown entertainers say 'It' is hurting their business",
 244                 'description': 'Stephen King s horror film has much to answer for. Jelby and Mr Loopy the Clowns join us.',
 245                 'timestamp': 1505341164,
 246                 'upload_date': '20170913',
 247                 'uploader_id': '2376984109001',
 248             },
 249             'params': {
 250                 'skip_download': True,
 251             },
 252         },
 253         {
 254             # custom brightcove, geo-restricted to Australia, bypassable
 255             'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37263964/sunrise-episode-wed-27-sep/',
 256             'only_matching': True,
 257         }
 258     ]
 259
 260     def _real_extract(self, url):
 261         mobj = re.match(self._VALID_URL, url)
 262         page_id = mobj.group('id')
 263         display_id = mobj.group('display_id') or page_id
 264         host = mobj.group('host')
 265         webpage, urlh = self._download_webpage_handle(url, display_id)
 266         if 'err=404' in urlh.geturl():
 267             raise ExtractorError('Video gone', expected=True)
 268
 269         # Look for iframed media first
 270         entries = []
 271         iframe_urls = re.findall(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage)
 272         for idx, iframe_url in enumerate(iframe_urls):
 273             entries.append(self.url_result(host + iframe_url, 'Yahoo'))
 274         if entries:
 275             return self.playlist_result(entries, page_id)
 276
 277         # Look for NBCSports iframes
 278         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
 279         if nbc_sports_url:
 280             return self.url_result(nbc_sports_url, NBCSportsVPlayerIE.ie_key())
 281
 282         # Look for Brightcove Legacy Studio embeds
 283         bc_url = BrightcoveLegacyIE._extract_brightcove_url(webpage)
 284         if bc_url:
 285             return self.url_result(bc_url, BrightcoveLegacyIE.ie_key())
 286
 287         def brightcove_url_result(bc_url):
 288             return self.url_result(
 289                 smuggle_url(bc_url, {'geo_countries': [mobj.group('country')]}),
 290                 BrightcoveNewIE.ie_key())
 291
 292         # Look for Brightcove New Studio embeds
 293         bc_url = BrightcoveNewIE._extract_url(self, webpage)
 294         if bc_url:
 295             return brightcove_url_result(bc_url)
 296
 297         brightcove_iframe = self._search_regex(
 298             r'(<iframe[^>]+data-video-id=["\']\d+[^>]+>)', webpage,
 299             'brightcove iframe', default=None)
 300         if brightcove_iframe:
 301             attr = extract_attributes(brightcove_iframe)
 302             src = attr.get('src')
 303             if src:
 304                 parsed_src = compat_urlparse.urlparse(src)
 305                 qs = compat_urlparse.parse_qs(parsed_src.query)
 306                 account_id = qs.get('accountId', ['2376984109001'])[0]
 307                 brightcove_id = attr.get('data-video-id') or qs.get('videoId', [None])[0]
 308                 if account_id and brightcove_id:
 309                     return brightcove_url_result(
 310                         'http://players.brightcove.net/%s/default_default/index.html?videoId=%s'
 311                         % (account_id, brightcove_id))
 312
 313         # Query result is often embedded in webpage as JSON. Sometimes explicit requests
 314         # to video API results in a failure with geo restriction reason therefore using
 315         # embedded query result when present sounds reasonable.
 316         config_json = self._search_regex(
 317             r'window\.Af\.bootstrap\[[^\]]+\]\s*=\s*({.*?"applet_type"\s*:\s*"td-applet-videoplayer".*?});(?:</script>|$)',
 318             webpage, 'videoplayer applet', default=None)
 319         if config_json:
 320             config = self._parse_json(config_json, display_id, fatal=False)
 321             if config:
 322                 sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi')
 323                 if sapi and 'query' in sapi:
 324                     info = self._extract_info(display_id, sapi, webpage)
 325                     self._sort_formats(info['formats'])
 326                     return info
 327
 328         items_json = self._search_regex(
 329             r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
 330             default=None)
 331         if items_json is None:
 332             alias = self._search_regex(
 333                 r'"aliases":{"video":"(.*?)"', webpage, 'alias', default=None)
 334             if alias is not None:
 335                 alias_info = self._download_json(
 336                     'https://www.yahoo.com/_td/api/resource/VideoService.videos;video_aliases=["%s"]' % alias,
 337                     display_id, 'Downloading alias info')
 338                 video_id = alias_info[0]['id']
 339             else:
 340                 CONTENT_ID_REGEXES = [
 341                     r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
 342                     r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"',
 343                     r'"first_videoid"\s*:\s*"([^"]+)"',
 344                     r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id),
 345                     r'<article[^>]data-uuid=["\']([^"\']+)',
 346                     r'<meta[^<>]+yahoo://article/view\?.*\buuid=([^&"\']+)',
 347                     r'<meta[^<>]+["\']ytwnews://cavideo/(?:[^/]+/)+([\da-fA-F-]+)[&"\']',
 348                 ]
 349                 video_id = self._search_regex(
 350                     CONTENT_ID_REGEXES, webpage, 'content ID')
 351         else:
 352             items = json.loads(items_json)
 353             info = items['mediaItems']['query']['results']['mediaObj'][0]
 354             # The 'meta' field is not always in the video webpage, we request it
 355             # from another page
 356             video_id = info['id']
 357         return self._get_info(video_id, display_id, webpage)
 358
 359     def _extract_info(self, display_id, query, webpage):
 360         info = query['query']['results']['mediaObj'][0]
 361         meta = info.get('meta')
 362         video_id = info.get('id')
 363
 364         if not meta:
 365             msg = info['status'].get('msg')
 366             if msg:
 367                 raise ExtractorError(
 368                     '%s returned error: %s' % (self.IE_NAME, msg), expected=True)
 369             raise ExtractorError('Unable to extract media object meta')
 370
 371         formats = []
 372         for s in info['streams']:
 373             tbr = int_or_none(s.get('bitrate'))
 374             format_info = {
 375                 'width': int_or_none(s.get('width')),
 376                 'height': int_or_none(s.get('height')),
 377                 'tbr': tbr,
 378             }
 379
 380             host = s['host']
 381             path = s['path']
 382             if host.startswith('rtmp'):
 383                 fmt = 'rtmp'
 384                 format_info.update({
 385                     'url': host,
 386                     'play_path': path,
 387                     'ext': 'flv',
 388                 })
 389             else:
 390                 if s.get('format') == 'm3u8_playlist':
 391                     fmt = 'hls'
 392                     format_info.update({
 393                         'protocol': 'm3u8_native',
 394                         'ext': 'mp4',
 395                     })
 396                 else:
 397                     fmt = format_info['ext'] = determine_ext(path)
 398                 format_url = compat_urlparse.urljoin(host, path)
 399                 format_info['url'] = format_url
 400             format_info['format_id'] = fmt + ('-%d' % tbr if tbr else '')
 401             formats.append(format_info)
 402
 403         closed_captions = self._html_search_regex(
 404             r'"closedcaptions":(\[[^\]]+\])', webpage, 'closed captions',
 405             default='[]')
 406
 407         cc_json = self._parse_json(closed_captions, video_id, fatal=False)
 408         subtitles = {}
 409         if cc_json:
 410             for closed_caption in cc_json:
 411                 lang = closed_caption['lang']
 412                 if lang not in subtitles:
 413                     subtitles[lang] = []
 414                 subtitles[lang].append({
 415                     'url': closed_caption['url'],
 416                     'ext': mimetype2ext(closed_caption['content_type']),
 417                 })
 418
 419         return {
 420             'id': video_id,
 421             'display_id': display_id,
 422             'title': unescapeHTML(meta['title']),
 423             'formats': formats,
 424             'description': clean_html(meta['description']),
 425             'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
 426             'duration': int_or_none(meta.get('duration')),
 427             'subtitles': subtitles,
 428         }
 429
 430     def _get_info(self, video_id, display_id, webpage):
 431         region = self._search_regex(
 432             r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"',
 433             webpage, 'region', fatal=False, default='US').upper()
 434         formats = []
 435         info = {}
 436         for fmt in ('webm', 'mp4'):
 437             query_result = self._download_json(
 438                 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' + video_id,
 439                 display_id, 'Downloading %s video info' % fmt, query={
 440                     'protocol': 'http',
 441                     'region': region,
 442                     'format': fmt,
 443                 })
 444             info = self._extract_info(display_id, query_result, webpage)
 445             formats.extend(info['formats'])
 446         formats.extend(self._extract_m3u8_formats(
 447             'http://video.media.yql.yahoo.com/v1/hls/%s?region=%s' % (video_id, region),
 448             video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
 449         self._sort_formats(formats)
 450         info['formats'] = formats
 451         return info
 452
 453
 454 class YahooSearchIE(SearchInfoExtractor):
 455     IE_DESC = 'Yahoo screen search'
 456     _MAX_RESULTS = 1000
 457     IE_NAME = 'screen.yahoo:search'
 458     _SEARCH_KEY = 'yvsearch'
 459
 460     def _get_n_results(self, query, n):
 461         """Get a specified number of results for a query"""
 462         entries = []
 463         for pagenum in itertools.count(0):
 464             result_url = 'http://video.search.yahoo.com/search/?p=%s&fr=screen&o=js&gs=0&b=%d' % (compat_urllib_parse.quote_plus(query), pagenum * 30)
 465             info = self._download_json(result_url, query,
 466                                        note='Downloading results page ' + str(pagenum + 1))
 467             m = info['m']
 468             results = info['results']
 469
 470             for (i, r) in enumerate(results):
 471                 if (pagenum * 30) + i >= n:
 472                     break
 473                 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
 474                 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
 475                 entries.append(e)
 476             if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):
 477                 break
 478
 479         return {
 480             '_type': 'playlist',
 481             'id': query,
 482             'entries': entries,
 483         }
 484
 485
 486 class YahooGyaOPlayerIE(InfoExtractor):
 487     IE_NAME = 'yahoo:gyao:player'
 488     _VALID_URL = r'https?://(?:gyao\.yahoo\.co\.jp/(?:player|episode/[^/]+)|streaming\.yahoo\.co\.jp/c/y)/(?P<id>\d+/v\d+/v\d+|[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
 489     _TESTS = [{
 490         'url': 'https://gyao.yahoo.co.jp/player/00998/v00818/v0000000000000008564/',
 491         'info_dict': {
 492             'id': '5993125228001',
 493             'ext': 'mp4',
 494             'title': 'フューリー　【字幕版】',
 495             'description': 'md5:21e691c798a15330eda4db17a8fe45a5',
 496             'uploader_id': '4235717419001',
 497             'upload_date': '20190124',
 498             'timestamp': 1548294365,
 499         },
 500         'params': {
 501             # m3u8 download
 502             'skip_download': True,
 503         },
 504     }, {
 505         'url': 'https://streaming.yahoo.co.jp/c/y/01034/v00133/v0000000000000000706/',
 506         'only_matching': True,
 507     }, {
 508         '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',
 509         'only_matching': True,
 510     }]
 511
 512     def _real_extract(self, url):
 513         video_id = self._match_id(url).replace('/', ':')
 514         video = self._download_json(
 515             'https://gyao.yahoo.co.jp/dam/v1/videos/' + video_id,
 516             video_id, query={
 517                 'fields': 'longDescription,title,videoId',
 518             }, headers={
 519                 'X-User-Agent': 'Unknown Pc GYAO!/2.0.0 Web',
 520             })
 521         return {
 522             '_type': 'url_transparent',
 523             'id': video_id,
 524             'title': video['title'],
 525             'url': smuggle_url(
 526                 'http://players.brightcove.net/4235717419001/default_default/index.html?videoId=' + video['videoId'],
 527                 {'geo_countries': ['JP']}),
 528             'description': video.get('longDescription'),
 529             'ie_key': BrightcoveNewIE.ie_key(),
 530         }
 531
 532
 533 class YahooGyaOIE(InfoExtractor):
 534     IE_NAME = 'yahoo:gyao'
 535     _VALID_URL = r'https?://(?:gyao\.yahoo\.co\.jp/(?:p|title/[^/]+)|streaming\.yahoo\.co\.jp/p/y)/(?P<id>\d+/v\d+|[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
 536     _TESTS = [{
 537         'url': 'https://gyao.yahoo.co.jp/p/00449/v03102/',
 538         'info_dict': {
 539             'id': '00449:v03102',
 540         },
 541         'playlist_count': 2,
 542     }, {
 543         'url': 'https://streaming.yahoo.co.jp/p/y/01034/v00133/',
 544         'only_matching': True,
 545     }, {
 546         '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',
 547         'only_matching': True,
 548     }]
 549
 550     def _real_extract(self, url):
 551         program_id = self._match_id(url).replace('/', ':')
 552         videos = self._download_json(
 553             'https://gyao.yahoo.co.jp/api/programs/%s/videos' % program_id, program_id)['videos']
 554         entries = []
 555         for video in videos:
 556             video_id = video.get('id')
 557             if not video_id:
 558                 continue
 559             entries.append(self.url_result(
 560                 'https://gyao.yahoo.co.jp/player/%s/' % video_id.replace(':', '/'),
 561                 YahooGyaOPlayerIE.ie_key(), video_id))
 562         return self.playlist_result(entries, program_id)
 563
 564
 565 class YahooJapanNewsIE(InfoExtractor):
 566     IE_NAME = 'yahoo:japannews'
 567     IE_DESC = 'Yahoo! Japan News'
 568     _VALID_URL = r'https?://(?P<host>(?:news|headlines)\.yahoo\.co\.jp)[^\d]*(?P<id>\d[\d-]*\d)?'
 569     _GEO_COUNTRIES = ['JP']
 570     _TESTS = [{
 571         'url': 'https://headlines.yahoo.co.jp/videonews/ann?a=20190716-00000071-ann-int',
 572         'info_dict': {
 573             'id': '1736242',
 574             'ext': 'mp4',
 575             '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',
 576             '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）)',
 577             'thumbnail': r're:^https?://.*\.[a-zA-Z\d]{3,4}$',
 578         },
 579         'params': {
 580             'skip_download': True,
 581         },
 582     }, {
 583         # geo restricted
 584         'url': 'https://headlines.yahoo.co.jp/hl?a=20190721-00000001-oxv-l04',
 585         'only_matching': True,
 586     }, {
 587         'url': 'https://headlines.yahoo.co.jp/videonews/',
 588         'only_matching': True,
 589     }, {
 590         'url': 'https://news.yahoo.co.jp',
 591         'only_matching': True,
 592     }, {
 593         'url': 'https://news.yahoo.co.jp/byline/hashimotojunji/20190628-00131977/',
 594         'only_matching': True,
 595     }, {
 596         'url': 'https://news.yahoo.co.jp/feature/1356',
 597         'only_matching': True
 598     }]
 599
 600     def _extract_formats(self, json_data, content_id):
 601         formats = []
 602
 603         video_data = try_get(
 604             json_data,
 605             lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'],
 606             list)
 607         for vid in video_data or []:
 608             delivery = vid.get('delivery')
 609             url = url_or_none(vid.get('Url'))
 610             if not delivery or not url:
 611                 continue
 612             elif delivery == 'hls':
 613                 formats.extend(
 614                     self._extract_m3u8_formats(
 615                         url, content_id, 'mp4', 'm3u8_native',
 616                         m3u8_id='hls', fatal=False))
 617             else:
 618                 formats.append({
 619                     'url': url,
 620                     'format_id': 'http-%s' % compat_str(vid.get('bitrate', '')),
 621                     'height': int_or_none(vid.get('height')),
 622                     'width': int_or_none(vid.get('width')),
 623                     'tbr': int_or_none(vid.get('bitrate')),
 624                 })
 625         self._remove_duplicate_formats(formats)
 626         self._sort_formats(formats)
 627
 628         return formats
 629
 630     def _real_extract(self, url):
 631         mobj = re.match(self._VALID_URL, url)
 632         host = mobj.group('host')
 633         display_id = mobj.group('id') or host
 634
 635         webpage = self._download_webpage(url, display_id)
 636
 637         title = self._html_search_meta(
 638             ['og:title', 'twitter:title'], webpage, 'title', default=None
 639         ) or self._html_search_regex('<title>([^<]+)</title>', webpage, 'title')
 640
 641         if display_id == host:
 642             # Headline page (w/ multiple BC playlists) ('news.yahoo.co.jp', 'headlines.yahoo.co.jp/videonews/', ...)
 643             stream_plists = re.findall(r'plist=(\d+)', webpage) or re.findall(r'plist["\']:\s*["\']([^"\']+)', webpage)
 644             entries = [
 645                 self.url_result(
 646                     smuggle_url(
 647                         'http://players.brightcove.net/5690807595001/HyZNerRl7_default/index.html?playlistId=%s' % plist_id,
 648                         {'geo_countries': ['JP']}),
 649                     ie='BrightcoveNew', video_id=plist_id)
 650                 for plist_id in stream_plists]
 651             return self.playlist_result(entries, playlist_title=title)
 652
 653         # Article page
 654         description = self._html_search_meta(
 655             ['og:description', 'description', 'twitter:description'],
 656             webpage, 'description', default=None)
 657         thumbnail = self._og_search_thumbnail(
 658             webpage, default=None) or self._html_search_meta(
 659             'twitter:image', webpage, 'thumbnail', default=None)
 660         space_id = self._search_regex([
 661             r'<script[^>]+class=["\']yvpub-player["\'][^>]+spaceid=([^&"\']+)',
 662             r'YAHOO\.JP\.srch\.\w+link\.onLoad[^;]+spaceID["\' ]*:["\' ]+([^"\']+)',
 663             r'<!--\s+SpaceID=(\d+)'
 664         ], webpage, 'spaceid')
 665
 666         content_id = self._search_regex(
 667             r'<script[^>]+class=["\']yvpub-player["\'][^>]+contentid=(?P<contentid>[^&"\']+)',
 668             webpage, 'contentid', group='contentid')
 669
 670         json_data = self._download_json(
 671             'https://feapi-yvpub.yahooapis.jp/v1/content/%s' % content_id,
 672             content_id,
 673             query={
 674                 'appid': 'dj0zaiZpPVZMTVFJR0FwZWpiMyZzPWNvbnN1bWVyc2VjcmV0Jng9YjU-',
 675                 'output': 'json',
 676                 'space_id': space_id,
 677                 'domain': host,
 678                 'ak': hashlib.md5('_'.join((space_id, host)).encode()).hexdigest(),
 679                 'device_type': '1100',
 680             })
 681         formats = self._extract_formats(json_data, content_id)
 682
 683         return {
 684             'id': content_id,
 685             'title': title,
 686             'description': description,
 687             'thumbnail': thumbnail,
 688             'formats': formats,
 689         }