_ Git - youtube-dl/blob - youtube_dl/extractor/generic.py

   1 # encoding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 import os
   6 import re
   7
   8 from .common import InfoExtractor
   9 from .youtube import YoutubeIE
  10 from ..utils import (
  11     compat_urllib_parse,
  12     compat_urlparse,
  13     compat_xml_parse_error,
  14
  15     determine_ext,
  16     ExtractorError,
  17     float_or_none,
  18     HEADRequest,
  19     orderedSet,
  20     parse_xml,
  21     smuggle_url,
  22     unescapeHTML,
  23     unified_strdate,
  24     unsmuggle_url,
  25     url_basename,
  26 )
  27 from .brightcove import BrightcoveIE
  28 from .ooyala import OoyalaIE
  29 from .rutv import RUTVIE
  30 from .smotri import SmotriIE
  31
  32
  33 class GenericIE(InfoExtractor):
  34     IE_DESC = 'Generic downloader that works on some sites'
  35     _VALID_URL = r'.*'
  36     IE_NAME = 'generic'
  37     _TESTS = [
  38         {
  39             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
  40             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
  41             'info_dict': {
  42                 'id': '13601338388002',
  43                 'ext': 'mp4',
  44                 'uploader': 'www.hodiho.fr',
  45                 'title': 'R\u00e9gis plante sa Jeep',
  46             }
  47         },
  48         # bandcamp page with custom domain
  49         {
  50             'add_ie': ['Bandcamp'],
  51             'url': 'http://bronyrock.com/track/the-pony-mash',
  52             'info_dict': {
  53                 'id': '3235767654',
  54                 'ext': 'mp3',
  55                 'title': 'The Pony Mash',
  56                 'uploader': 'M_Pallante',
  57             },
  58             'skip': 'There is a limit of 200 free downloads / month for the test song',
  59         },
  60         # embedded brightcove video
  61         # it also tests brightcove videos that need to set the 'Referer' in the
  62         # http requests
  63         {
  64             'add_ie': ['Brightcove'],
  65             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
  66             'info_dict': {
  67                 'id': '2765128793001',
  68                 'ext': 'mp4',
  69                 'title': 'Le cours de bourse : l’analyse technique',
  70                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
  71                 'uploader': 'BFM BUSINESS',
  72             },
  73             'params': {
  74                 'skip_download': True,
  75             },
  76         },
  77         {
  78             # https://github.com/rg3/youtube-dl/issues/2253
  79             'url': 'http://bcove.me/i6nfkrc3',
  80             'md5': '0ba9446db037002366bab3b3eb30c88c',
  81             'info_dict': {
  82                 'id': '3101154703001',
  83                 'ext': 'mp4',
  84                 'title': 'Still no power',
  85                 'uploader': 'thestar.com',
  86                 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.',
  87             },
  88             'add_ie': ['Brightcove'],
  89         },
  90         {
  91             'url': 'http://www.championat.com/video/football/v/87/87499.html',
  92             'md5': 'fb973ecf6e4a78a67453647444222983',
  93             'info_dict': {
  94                 'id': '3414141473001',
  95                 'ext': 'mp4',
  96                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
  97                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
  98                 'uploader': 'Championat',
  99             },
 100         },
 101         # Direct link to a video
 102         {
 103             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
 104             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
 105             'info_dict': {
 106                 'id': 'trailer',
 107                 'ext': 'mp4',
 108                 'title': 'trailer',
 109                 'upload_date': '20100513',
 110             }
 111         },
 112         # ooyala video
 113         {
 114             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 115             'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
 116             'info_dict': {
 117                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 118                 'ext': 'mp4',
 119                 'title': '2cc213299525360.mov',  # that's what we get
 120             },
 121         },
 122         # google redirect
 123         {
 124             'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
 125             'info_dict': {
 126                 'id': 'cmQHVoWB5FY',
 127                 'ext': 'mp4',
 128                 'upload_date': '20130224',
 129                 'uploader_id': 'TheVerge',
 130                 'description': 'Chris Ziegler takes a look at the Alcatel OneTouch Fire and the ZTE Open; two of the first Firefox OS handsets to be officially announced.',
 131                 'uploader': 'The Verge',
 132                 'title': 'First Firefox OS phones side-by-side',
 133             },
 134             'params': {
 135                 'skip_download': False,
 136             }
 137         },
 138         # embed.ly video
 139         {
 140             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 141             'info_dict': {
 142                 'id': '9ODmcdjQcHQ',
 143                 'ext': 'mp4',
 144                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 145                 'upload_date': '20140225',
 146                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 147                 'uploader': 'Tested',
 148                 'uploader_id': 'testedcom',
 149             },
 150             # No need to test YoutubeIE here
 151             'params': {
 152                 'skip_download': True,
 153             },
 154         },
 155         # funnyordie embed
 156         {
 157             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 158             'info_dict': {
 159                 'id': '18e820ec3f',
 160                 'ext': 'mp4',
 161                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 162                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 163             },
 164         },
 165         # RUTV embed
 166         {
 167             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 168             'info_dict': {
 169                 'id': '776940',
 170                 'ext': 'mp4',
 171                 'title': 'Охотское море стало целиком российским',
 172                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 173             },
 174             'params': {
 175                 # m3u8 download
 176                 'skip_download': True,
 177             },
 178         },
 179         # Embedded TED video
 180         {
 181             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 182             'md5': '65fdff94098e4a607385a60c5177c638',
 183             'info_dict': {
 184                 'id': '1969',
 185                 'ext': 'mp4',
 186                 'title': 'Hidden miracles of the natural world',
 187                 'uploader': 'Louie Schwartzberg',
 188                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 189             }
 190         },
 191         # Embeded Ustream video
 192         {
 193             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 194             'md5': '27b99cdb639c9b12a79bca876a073417',
 195             'info_dict': {
 196                 'id': '45734260',
 197                 'ext': 'flv',
 198                 'uploader': 'AU SPA:  The NSA and Privacy',
 199                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 200             }
 201         },
 202         # nowvideo embed hidden behind percent encoding
 203         {
 204             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 205             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 206             'info_dict': {
 207                 'id': '06e53103ca9aa',
 208                 'ext': 'flv',
 209                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 210                 'description': 'No description',
 211             },
 212         },
 213         # arte embed
 214         {
 215             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 216             'md5': '7653032cbb25bf6c80d80f217055fa43',
 217             'info_dict': {
 218                 'id': '048195-004_PLUS7-F',
 219                 'ext': 'flv',
 220                 'title': 'X:enius',
 221                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 222                 'upload_date': '20140320',
 223             },
 224             'params': {
 225                 'skip_download': 'Requires rtmpdump'
 226             }
 227         },
 228         # smotri embed
 229         {
 230             'url': 'http://rbctv.rbc.ru/archive/news/562949990879132.shtml',
 231             'md5': 'ec40048448e9284c9a1de77bb188108b',
 232             'info_dict': {
 233                 'id': 'v27008541fad',
 234                 'ext': 'mp4',
 235                 'title': 'Крым и Севастополь вошли в состав России',
 236                 'description': 'md5:fae01b61f68984c7bd2fa741e11c3175',
 237                 'duration': 900,
 238                 'upload_date': '20140318',
 239                 'uploader': 'rbctv_2012_4',
 240                 'uploader_id': 'rbctv_2012_4',
 241             },
 242         },
 243         # Condé Nast embed
 244         {
 245             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 246             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 247             'info_dict': {
 248                 'id': '53501be369702d3275860000',
 249                 'ext': 'mp4',
 250                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 251             }
 252         },
 253         # Dailymotion embed
 254         {
 255             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 256             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 257             'info_dict': {
 258                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 259                 'ext': 'mp4',
 260                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 261                 'uploader': 'Spi0n',
 262             },
 263             'add_ie': ['Dailymotion'],
 264         },
 265         # YouTube embed
 266         {
 267             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 268             'info_dict': {
 269                 'id': 'FXRb4ykk4S0',
 270                 'ext': 'mp4',
 271                 'title': 'The NBL Auction 2014',
 272                 'uploader': 'BADMINTON England',
 273                 'uploader_id': 'BADMINTONEvents',
 274                 'upload_date': '20140603',
 275                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 276             },
 277             'add_ie': ['Youtube'],
 278             'params': {
 279                 'skip_download': True,
 280             }
 281         },
 282         # MTVSercices embed
 283         {
 284             'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
 285             'md5': '35727f82f58c76d996fc188f9755b0d5',
 286             'info_dict': {
 287                 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
 288                 'ext': 'mp4',
 289                 'title': 'Review',
 290                 'description': 'Mario\'s life in the fast lane has never looked so good.',
 291             },
 292         },
 293         # YouTube embed via <data-embed-url="">
 294         {
 295             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 296             'info_dict': {
 297                 'id': '4vAffPZIT44',
 298                 'ext': 'mp4',
 299                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 300                 'uploader': 'Gameloft',
 301                 'uploader_id': 'gameloft',
 302                 'upload_date': '20140828',
 303                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 304             },
 305             'params': {
 306                 'skip_download': True,
 307             }
 308         },
 309         # Camtasia studio
 310         {
 311             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 312             'playlist': [{
 313                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 314                 'info_dict': {
 315                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 316                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 317                     'ext': 'flv',
 318                     'duration': 2235.90,
 319                 }
 320             }, {
 321                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 322                 'info_dict': {
 323                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 324                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 325                     'ext': 'flv',
 326                     'duration': 2235.93,
 327                 }
 328             }],
 329             'info_dict': {
 330                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 331             }
 332         },
 333         # Flowplayer
 334         {
 335             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 336             'md5': '9d65602bf31c6e20014319c7d07fba27',
 337             'info_dict': {
 338                 'id': '5123ea6d5e5a7',
 339                 'ext': 'mp4',
 340                 'age_limit': 18,
 341                 'uploader': 'www.handjobhub.com',
 342                 'title': 'Busty Blonde Siri Tit Fuck While Wank at Handjob Hub',
 343             }
 344         },
 345         # RSS feed
 346         {
 347             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 348             'info_dict': {
 349                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 350                 'title': 'Zero Punctuation',
 351                 'description': 're:'
 352             },
 353             'playlist_mincount': 11,
 354         },
 355         # Multiple brightcove videos
 356         # https://github.com/rg3/youtube-dl/issues/2283
 357         {
 358             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 359             'info_dict': {
 360                 'id': 'always-never',
 361                 'title': 'Always / Never - The New Yorker',
 362             },
 363             'playlist_count': 3,
 364             'params': {
 365                 'extract_flat': False,
 366                 'skip_download': True,
 367             }
 368         },
 369         # MLB embed
 370         {
 371             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 372             'md5': '96f09a37e44da40dd083e12d9a683327',
 373             'info_dict': {
 374                 'id': '33322633',
 375                 'ext': 'mp4',
 376                 'title': 'Ump changes call to ball',
 377                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 378                 'duration': 48,
 379                 'timestamp': 1401537900,
 380                 'upload_date': '20140531',
 381                 'thumbnail': 're:^https?://.*\.jpg$',
 382             },
 383         },
 384         # Wistia embed
 385         {
 386             'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 387             'md5': '8788b683c777a5cf25621eaf286d0c23',
 388             'info_dict': {
 389                 'id': '1cfaf6b7ea',
 390                 'ext': 'mov',
 391                 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
 392                 'duration': 643.0,
 393                 'filesize': 182808282,
 394                 'uploader': 'education-portal.com',
 395             },
 396         },
 397     ]
 398
 399     def report_following_redirect(self, new_url):
 400         """Report information extraction."""
 401         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
 402
 403     def _extract_rss(self, url, video_id, doc):
 404         playlist_title = doc.find('./channel/title').text
 405         playlist_desc_el = doc.find('./channel/description')
 406         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
 407
 408         entries = [{
 409             '_type': 'url',
 410             'url': e.find('link').text,
 411             'title': e.find('title').text,
 412         } for e in doc.findall('./channel/item')]
 413
 414         return {
 415             '_type': 'playlist',
 416             'id': url,
 417             'title': playlist_title,
 418             'description': playlist_desc,
 419             'entries': entries,
 420         }
 421
 422     def _extract_camtasia(self, url, video_id, webpage):
 423         """ Returns None if no camtasia video can be found. """
 424
 425         camtasia_cfg = self._search_regex(
 426             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
 427             webpage, 'camtasia configuration file', default=None)
 428         if camtasia_cfg is None:
 429             return None
 430
 431         title = self._html_search_meta('DC.title', webpage, fatal=True)
 432
 433         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
 434         camtasia_cfg = self._download_xml(
 435             camtasia_url, video_id,
 436             note='Downloading camtasia configuration',
 437             errnote='Failed to download camtasia configuration')
 438         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
 439
 440         entries = []
 441         for n in fileset_node.getchildren():
 442             url_n = n.find('./uri')
 443             if url_n is None:
 444                 continue
 445
 446             entries.append({
 447                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
 448                 'title': '%s - %s' % (title, n.tag),
 449                 'url': compat_urlparse.urljoin(url, url_n.text),
 450                 'duration': float_or_none(n.find('./duration').text),
 451             })
 452
 453         return {
 454             '_type': 'playlist',
 455             'entries': entries,
 456             'title': title,
 457         }
 458
 459     def _real_extract(self, url):
 460         if url.startswith('//'):
 461             return {
 462                 '_type': 'url',
 463                 'url': self.http_scheme() + url,
 464             }
 465
 466         parsed_url = compat_urlparse.urlparse(url)
 467         if not parsed_url.scheme:
 468             default_search = self._downloader.params.get('default_search')
 469             if default_search is None:
 470                 default_search = 'fixup_error'
 471
 472             if default_search in ('auto', 'auto_warning', 'fixup_error'):
 473                 if '/' in url:
 474                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
 475                     return self.url_result('http://' + url)
 476                 elif default_search != 'fixup_error':
 477                     if default_search == 'auto_warning':
 478                         if re.match(r'^(?:url|URL)$', url):
 479                             raise ExtractorError(
 480                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
 481                                 expected=True)
 482                         else:
 483                             self._downloader.report_warning(
 484                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
 485                     return self.url_result('ytsearch:' + url)
 486
 487             if default_search in ('error', 'fixup_error'):
 488                 raise ExtractorError(
 489                     ('%r is not a valid URL. '
 490                      'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
 491                     ) % (url, url), expected=True)
 492             else:
 493                 assert ':' in default_search
 494                 return self.url_result(default_search + url)
 495
 496         url, smuggled_data = unsmuggle_url(url)
 497         force_videoid = None
 498         is_intentional = smuggled_data and smuggled_data.get('to_generic')
 499         if smuggled_data and 'force_videoid' in smuggled_data:
 500             force_videoid = smuggled_data['force_videoid']
 501             video_id = force_videoid
 502         else:
 503             video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
 504
 505         self.to_screen('%s: Requesting header' % video_id)
 506
 507         head_req = HEADRequest(url)
 508         response = self._request_webpage(
 509             head_req, video_id,
 510             note=False, errnote='Could not send HEAD request to %s' % url,
 511             fatal=False)
 512
 513         if response is not False:
 514             # Check for redirect
 515             new_url = response.geturl()
 516             if url != new_url:
 517                 self.report_following_redirect(new_url)
 518                 if force_videoid:
 519                     new_url = smuggle_url(
 520                         new_url, {'force_videoid': force_videoid})
 521                 return self.url_result(new_url)
 522
 523             # Check for direct link to a video
 524             content_type = response.headers.get('Content-Type', '')
 525             m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
 526             if m:
 527                 upload_date = response.headers.get('Last-Modified')
 528                 if upload_date:
 529                     upload_date = unified_strdate(upload_date)
 530                 return {
 531                     'id': video_id,
 532                     'title': os.path.splitext(url_basename(url))[0],
 533                     'formats': [{
 534                         'format_id': m.group('format_id'),
 535                         'url': url,
 536                         'vcodec': 'none' if m.group('type') == 'audio' else None
 537                     }],
 538                     'upload_date': upload_date,
 539                 }
 540
 541         if not self._downloader.params.get('test', False) and not is_intentional:
 542             self._downloader.report_warning('Falling back on generic information extractor.')
 543
 544         try:
 545             webpage = self._download_webpage(url, video_id)
 546         except ValueError:
 547             # since this is the last-resort InfoExtractor, if
 548             # this error is thrown, it'll be thrown here
 549             raise ExtractorError('Failed to download URL: %s' % url)
 550
 551         self.report_extraction(video_id)
 552
 553         # Is it an RSS feed?
 554         try:
 555             doc = parse_xml(webpage)
 556             if doc.tag == 'rss':
 557                 return self._extract_rss(url, video_id, doc)
 558         except compat_xml_parse_error:
 559             pass
 560
 561         # Is it a Camtasia project?
 562         camtasia_res = self._extract_camtasia(url, video_id, webpage)
 563         if camtasia_res is not None:
 564             return camtasia_res
 565
 566         # Sometimes embedded video player is hidden behind percent encoding
 567         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
 568         # Unescaping the whole page allows to handle those cases in a generic way
 569         webpage = compat_urllib_parse.unquote(webpage)
 570
 571         # it's tempting to parse this further, but you would
 572         # have to take into account all the variations like
 573         #   Video Title - Site Name
 574         #   Site Name | Video Title
 575         #   Video Title - Tagline | Site Name
 576         # and so on and so forth; it's just not practical
 577         video_title = self._html_search_regex(
 578             r'(?s)<title>(.*?)</title>', webpage, 'video title',
 579             default='video')
 580
 581         # Try to detect age limit automatically
 582         age_limit = self._rta_search(webpage)
 583         # And then there are the jokers who advertise that they use RTA,
 584         # but actually don't.
 585         AGE_LIMIT_MARKERS = [
 586             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
 587         ]
 588         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
 589             age_limit = 18
 590
 591         # video uploader is domain name
 592         video_uploader = self._search_regex(
 593             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
 594
 595         # Helper method
 596         def _playlist_from_matches(matches, getter, ie=None):
 597             urlrs = orderedSet(
 598                 self.url_result(self._proto_relative_url(getter(m)), ie)
 599                 for m in matches)
 600             return self.playlist_result(
 601                 urlrs, playlist_id=video_id, playlist_title=video_title)
 602
 603         # Look for BrightCove:
 604         bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
 605         if bc_urls:
 606             self.to_screen('Brightcove video detected.')
 607             entries = [{
 608                 '_type': 'url',
 609                 'url': smuggle_url(bc_url, {'Referer': url}),
 610                 'ie_key': 'Brightcove'
 611             } for bc_url in bc_urls]
 612
 613             return {
 614                 '_type': 'playlist',
 615                 'title': video_title,
 616                 'id': video_id,
 617                 'entries': entries,
 618             }
 619
 620         # Look for embedded (iframe) Vimeo player
 621         mobj = re.search(
 622             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
 623         if mobj:
 624             player_url = unescapeHTML(mobj.group('url'))
 625             surl = smuggle_url(player_url, {'Referer': url})
 626             return self.url_result(surl, 'Vimeo')
 627
 628         # Look for embedded (swf embed) Vimeo player
 629         mobj = re.search(
 630             r'<embed[^>]+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
 631         if mobj:
 632             return self.url_result(mobj.group(1), 'Vimeo')
 633
 634         # Look for embedded YouTube player
 635         matches = re.findall(r'''(?x)
 636             (?:
 637                 <iframe[^>]+?src=|
 638                 data-video-url=|
 639                 <embed[^>]+?src=|
 640                 embedSWF\(?:\s*
 641             )
 642             (["\'])
 643                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
 644                 (?:embed|v|p)/.+?)
 645             \1''', webpage)
 646         if matches:
 647             return _playlist_from_matches(
 648                 matches, lambda m: unescapeHTML(m[1]))
 649
 650         # Look for embedded Dailymotion player
 651         matches = re.findall(
 652             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
 653         if matches:
 654             return _playlist_from_matches(
 655                 matches, lambda m: unescapeHTML(m[1]))
 656
 657         # Look for embedded Wistia player
 658         match = re.search(
 659             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
 660         if match:
 661             return {
 662                 '_type': 'url_transparent',
 663                 'url': unescapeHTML(match.group('url')),
 664                 'ie_key': 'Wistia',
 665                 'uploader': video_uploader,
 666                 'title': video_title,
 667                 'id': video_id,
 668             }
 669         match = re.search(r'(?:id=["\']wistia_|data-wistiaid=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
 670         if match:
 671             return {
 672                 '_type': 'url_transparent',
 673                 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
 674                 'ie_key': 'Wistia',
 675                 'uploader': video_uploader,
 676                 'title': video_title,
 677                 'id': match.group('id')
 678             }
 679
 680         # Look for embedded blip.tv player
 681         mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
 682         if mobj:
 683             return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
 684         mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
 685         if mobj:
 686             return self.url_result(mobj.group(1), 'BlipTV')
 687
 688         # Look for embedded condenast player
 689         matches = re.findall(
 690             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
 691             webpage)
 692         if matches:
 693             return {
 694                 '_type': 'playlist',
 695                 'entries': [{
 696                     '_type': 'url',
 697                     'ie_key': 'CondeNast',
 698                     'url': ma,
 699                 } for ma in matches],
 700                 'title': video_title,
 701                 'id': video_id,
 702             }
 703
 704         # Look for Bandcamp pages with custom domain
 705         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
 706         if mobj is not None:
 707             burl = unescapeHTML(mobj.group(1))
 708             # Don't set the extractor because it can be a track url or an album
 709             return self.url_result(burl)
 710
 711         # Look for embedded Vevo player
 712         mobj = re.search(
 713             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
 714         if mobj is not None:
 715             return self.url_result(mobj.group('url'))
 716
 717         # Look for Ooyala videos
 718         mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
 719              re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
 720         if mobj is not None:
 721             return OoyalaIE._build_url_result(mobj.group('ec'))
 722
 723         # Look for Aparat videos
 724         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
 725         if mobj is not None:
 726             return self.url_result(mobj.group(1), 'Aparat')
 727
 728         # Look for MPORA videos
 729         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
 730         if mobj is not None:
 731             return self.url_result(mobj.group(1), 'Mpora')
 732
 733         # Look for embedded NovaMov-based player
 734         mobj = re.search(
 735             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
 736                     (?P<url>http://(?:(?:embed|www)\.)?
 737                         (?:novamov\.com|
 738                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
 739                            videoweed\.(?:es|com)|
 740                            movshare\.(?:net|sx|ag)|
 741                            divxstage\.(?:eu|net|ch|co|at|ag))
 742                         /embed\.php.+?)\1''', webpage)
 743         if mobj is not None:
 744             return self.url_result(mobj.group('url'))
 745
 746         # Look for embedded Facebook player
 747         mobj = re.search(
 748             r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
 749         if mobj is not None:
 750             return self.url_result(mobj.group('url'), 'Facebook')
 751
 752         # Look for embedded VK player
 753         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
 754         if mobj is not None:
 755             return self.url_result(mobj.group('url'), 'VK')
 756
 757         # Look for embedded ivi player
 758         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
 759         if mobj is not None:
 760             return self.url_result(mobj.group('url'), 'Ivi')
 761
 762         # Look for embedded Huffington Post player
 763         mobj = re.search(
 764             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
 765         if mobj is not None:
 766             return self.url_result(mobj.group('url'), 'HuffPost')
 767
 768         # Look for embed.ly
 769         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
 770         if mobj is not None:
 771             return self.url_result(mobj.group('url'))
 772         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
 773         if mobj is not None:
 774             return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
 775
 776         # Look for funnyordie embed
 777         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
 778         if matches:
 779             return _playlist_from_matches(
 780                 matches, getter=unescapeHTML, ie='FunnyOrDie')
 781
 782         # Look for embedded RUTV player
 783         rutv_url = RUTVIE._extract_url(webpage)
 784         if rutv_url:
 785             return self.url_result(rutv_url, 'RUTV')
 786
 787         # Look for embedded TED player
 788         mobj = re.search(
 789             r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
 790         if mobj is not None:
 791             return self.url_result(mobj.group('url'), 'TED')
 792
 793         # Look for embedded Ustream videos
 794         mobj = re.search(
 795             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
 796         if mobj is not None:
 797             return self.url_result(mobj.group('url'), 'Ustream')
 798
 799         # Look for embedded arte.tv player
 800         mobj = re.search(
 801             r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
 802             webpage)
 803         if mobj is not None:
 804             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
 805
 806         # Look for embedded smotri.com player
 807         smotri_url = SmotriIE._extract_url(webpage)
 808         if smotri_url:
 809             return self.url_result(smotri_url, 'Smotri')
 810
 811         # Look for embeded soundcloud player
 812         mobj = re.search(
 813             r'<iframe src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
 814             webpage)
 815         if mobj is not None:
 816             url = unescapeHTML(mobj.group('url'))
 817             return self.url_result(url)
 818
 819         # Look for embedded vulture.com player
 820         mobj = re.search(
 821             r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
 822             webpage)
 823         if mobj is not None:
 824             url = unescapeHTML(mobj.group('url'))
 825             return self.url_result(url, ie='Vulture')
 826
 827         # Look for embedded mtvservices player
 828         mobj = re.search(
 829             r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
 830             webpage)
 831         if mobj is not None:
 832             url = unescapeHTML(mobj.group('url'))
 833             return self.url_result(url, ie='MTVServicesEmbedded')
 834
 835         # Look for embedded yahoo player
 836         mobj = re.search(
 837             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
 838             webpage)
 839         if mobj is not None:
 840             return self.url_result(mobj.group('url'), 'Yahoo')
 841
 842         # Look for embedded sbs.com.au player
 843         mobj = re.search(
 844             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
 845             webpage)
 846         if mobj is not None:
 847             return self.url_result(mobj.group('url'), 'SBS')
 848
 849         mobj = re.search(
 850             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
 851             webpage)
 852         if mobj is not None:
 853             return self.url_result(mobj.group('url'), 'MLB')
 854
 855         # Start with something easy: JW Player in SWFObject
 856         found = re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
 857         if not found:
 858             # Look for gorilla-vid style embedding
 859             found = re.findall(r'''(?sx)
 860                 (?:
 861                     jw_plugins|
 862                     JWPlayerOptions|
 863                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
 864                 )
 865                 .*?file\s*:\s*["\'](.*?)["\']''', webpage)
 866         if not found:
 867             # Broaden the search a little bit
 868             found = re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
 869         if not found:
 870             # Broaden the findall a little bit: JWPlayer JS loader
 871             found = re.findall(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)
 872         if not found:
 873             # Flow player
 874             found = re.findall(r'''(?xs)
 875                 flowplayer\("[^"]+",\s*
 876                     \{[^}]+?\}\s*,
 877                     \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
 878                         ["']?url["']?\s*:\s*["']([^"']+)["']
 879             ''', webpage)
 880         if not found:
 881             # Try to find twitter cards info
 882             found = re.findall(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
 883         if not found:
 884             # We look for Open Graph info:
 885             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
 886             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
 887             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
 888             if m_video_type is not None:
 889                 def check_video(vurl):
 890                     vpath = compat_urlparse.urlparse(vurl).path
 891                     vext = determine_ext(vpath)
 892                     return '.' in vpath and vext not in ('swf', 'png', 'jpg')
 893                 found = list(filter(
 894                     check_video,
 895                     re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)))
 896         if not found:
 897             # HTML5 video
 898             found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]+)? src="([^"]+)"', webpage)
 899         if not found:
 900             found = re.search(
 901                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
 902                 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
 903                 webpage)
 904             if found:
 905                 new_url = found.group(1)
 906                 self.report_following_redirect(new_url)
 907                 return {
 908                     '_type': 'url',
 909                     'url': new_url,
 910                 }
 911         if not found:
 912             raise ExtractorError('Unsupported URL: %s' % url)
 913
 914         entries = []
 915         for video_url in found:
 916             video_url = compat_urlparse.urljoin(url, video_url)
 917             video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
 918
 919             # Sometimes, jwplayer extraction will result in a YouTube URL
 920             if YoutubeIE.suitable(video_url):
 921                 entries.append(self.url_result(video_url, 'Youtube'))
 922                 continue
 923
 924             # here's a fun little line of code for you:
 925             video_id = os.path.splitext(video_id)[0]
 926
 927             entries.append({
 928                 'id': video_id,
 929                 'url': video_url,
 930                 'uploader': video_uploader,
 931                 'title': video_title,
 932                 'age_limit': age_limit,
 933             })
 934
 935         if len(entries) == 1:
 936             return entries[0]
 937         else:
 938             for num, e in enumerate(entries, start=1):
 939                 e['title'] = '%s (%d)' % (e['title'], num)
 940             return {
 941                 '_type': 'playlist',
 942                 'entries': entries,
 943             }
 944