_ 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 ..compat import (
  11     compat_urllib_parse,
  12     compat_urlparse,
  13     compat_xml_parse_error,
  14 )
  15 from ..utils import (
  16     determine_ext,
  17     ExtractorError,
  18     float_or_none,
  19     HEADRequest,
  20     is_html,
  21     orderedSet,
  22     parse_xml,
  23     smuggle_url,
  24     unescapeHTML,
  25     unified_strdate,
  26     unsmuggle_url,
  27     UnsupportedError,
  28     url_basename,
  29     xpath_text,
  30 )
  31 from .brightcove import BrightcoveIE
  32 from .ooyala import OoyalaIE
  33 from .rutv import RUTVIE
  34 from .smotri import SmotriIE
  35 from .condenast import CondeNastIE
  36
  37
  38 class GenericIE(InfoExtractor):
  39     IE_DESC = 'Generic downloader that works on some sites'
  40     _VALID_URL = r'.*'
  41     IE_NAME = 'generic'
  42     _TESTS = [
  43         {
  44             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
  45             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
  46             'info_dict': {
  47                 'id': '13601338388002',
  48                 'ext': 'mp4',
  49                 'uploader': 'www.hodiho.fr',
  50                 'title': 'R\u00e9gis plante sa Jeep',
  51             }
  52         },
  53         # bandcamp page with custom domain
  54         {
  55             'add_ie': ['Bandcamp'],
  56             'url': 'http://bronyrock.com/track/the-pony-mash',
  57             'info_dict': {
  58                 'id': '3235767654',
  59                 'ext': 'mp3',
  60                 'title': 'The Pony Mash',
  61                 'uploader': 'M_Pallante',
  62             },
  63             'skip': 'There is a limit of 200 free downloads / month for the test song',
  64         },
  65         # embedded brightcove video
  66         # it also tests brightcove videos that need to set the 'Referer' in the
  67         # http requests
  68         {
  69             'add_ie': ['Brightcove'],
  70             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
  71             'info_dict': {
  72                 'id': '2765128793001',
  73                 'ext': 'mp4',
  74                 'title': 'Le cours de bourse : l’analyse technique',
  75                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
  76                 'uploader': 'BFM BUSINESS',
  77             },
  78             'params': {
  79                 'skip_download': True,
  80             },
  81         },
  82         {
  83             # https://github.com/rg3/youtube-dl/issues/2253
  84             'url': 'http://bcove.me/i6nfkrc3',
  85             'md5': '0ba9446db037002366bab3b3eb30c88c',
  86             'info_dict': {
  87                 'id': '3101154703001',
  88                 'ext': 'mp4',
  89                 'title': 'Still no power',
  90                 'uploader': 'thestar.com',
  91                 '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.',
  92             },
  93             'add_ie': ['Brightcove'],
  94         },
  95         {
  96             'url': 'http://www.championat.com/video/football/v/87/87499.html',
  97             'md5': 'fb973ecf6e4a78a67453647444222983',
  98             'info_dict': {
  99                 'id': '3414141473001',
 100                 'ext': 'mp4',
 101                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 102                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 103                 'uploader': 'Championat',
 104             },
 105         },
 106         {
 107             # https://github.com/rg3/youtube-dl/issues/3541
 108             'add_ie': ['Brightcove'],
 109             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 110             'info_dict': {
 111                 'id': '3866516442001',
 112                 'ext': 'mp4',
 113                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 114                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 115                 'uploader': 'SBS Broadcasting',
 116             },
 117             'skip': 'Restricted to Netherlands',
 118             'params': {
 119                 'skip_download': True,  # m3u8 download
 120             },
 121         },
 122         # Direct link to a video
 123         {
 124             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
 125             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
 126             'info_dict': {
 127                 'id': 'trailer',
 128                 'ext': 'mp4',
 129                 'title': 'trailer',
 130                 'upload_date': '20100513',
 131             }
 132         },
 133         # ooyala video
 134         {
 135             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 136             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 137             'info_dict': {
 138                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 139                 'ext': 'mp4',
 140                 'title': '2cc213299525360.mov',  # that's what we get
 141             },
 142             'add_ie': ['Ooyala'],
 143         },
 144         # multiple ooyala embeds on SBN network websites
 145         {
 146             'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 147             'info_dict': {
 148                 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 149                 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
 150             },
 151             'playlist_mincount': 3,
 152             'params': {
 153                 'skip_download': True,
 154             },
 155             'add_ie': ['Ooyala'],
 156         },
 157         # google redirect
 158         {
 159             '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',
 160             'info_dict': {
 161                 'id': 'cmQHVoWB5FY',
 162                 'ext': 'mp4',
 163                 'upload_date': '20130224',
 164                 'uploader_id': 'TheVerge',
 165                 'description': 're:^Chris Ziegler takes a look at the\.*',
 166                 'uploader': 'The Verge',
 167                 'title': 'First Firefox OS phones side-by-side',
 168             },
 169             'params': {
 170                 'skip_download': False,
 171             }
 172         },
 173         # embed.ly video
 174         {
 175             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 176             'info_dict': {
 177                 'id': '9ODmcdjQcHQ',
 178                 'ext': 'mp4',
 179                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 180                 'upload_date': '20140225',
 181                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 182                 'uploader': 'Tested',
 183                 'uploader_id': 'testedcom',
 184             },
 185             # No need to test YoutubeIE here
 186             'params': {
 187                 'skip_download': True,
 188             },
 189         },
 190         # funnyordie embed
 191         {
 192             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 193             'info_dict': {
 194                 'id': '18e820ec3f',
 195                 'ext': 'mp4',
 196                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 197                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 198             },
 199         },
 200         # BBC iPlayer embeds
 201         {
 202             'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
 203             'info_dict': {
 204                 'title': 'BBC - Blogs -  Adam Curtis - BUGGER',
 205             },
 206             'playlist_mincount': 18,
 207         },
 208         # RUTV embed
 209         {
 210             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 211             'info_dict': {
 212                 'id': '776940',
 213                 'ext': 'mp4',
 214                 'title': 'Охотское море стало целиком российским',
 215                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 216             },
 217             'params': {
 218                 # m3u8 download
 219                 'skip_download': True,
 220             },
 221         },
 222         # Embedded TED video
 223         {
 224             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 225             'md5': '65fdff94098e4a607385a60c5177c638',
 226             'info_dict': {
 227                 'id': '1969',
 228                 'ext': 'mp4',
 229                 'title': 'Hidden miracles of the natural world',
 230                 'uploader': 'Louie Schwartzberg',
 231                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 232             }
 233         },
 234         # Embeded Ustream video
 235         {
 236             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 237             'md5': '27b99cdb639c9b12a79bca876a073417',
 238             'info_dict': {
 239                 'id': '45734260',
 240                 'ext': 'flv',
 241                 'uploader': 'AU SPA:  The NSA and Privacy',
 242                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 243             }
 244         },
 245         # nowvideo embed hidden behind percent encoding
 246         {
 247             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 248             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 249             'info_dict': {
 250                 'id': '06e53103ca9aa',
 251                 'ext': 'flv',
 252                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 253                 'description': 'No description',
 254             },
 255         },
 256         # arte embed
 257         {
 258             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 259             'md5': '7653032cbb25bf6c80d80f217055fa43',
 260             'info_dict': {
 261                 'id': '048195-004_PLUS7-F',
 262                 'ext': 'flv',
 263                 'title': 'X:enius',
 264                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 265                 'upload_date': '20140320',
 266             },
 267             'params': {
 268                 'skip_download': 'Requires rtmpdump'
 269             }
 270         },
 271         # Condé Nast embed
 272         {
 273             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 274             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 275             'info_dict': {
 276                 'id': '53501be369702d3275860000',
 277                 'ext': 'mp4',
 278                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 279             }
 280         },
 281         # Dailymotion embed
 282         {
 283             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 284             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 285             'info_dict': {
 286                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 287                 'ext': 'mp4',
 288                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 289                 'uploader': 'Spi0n',
 290             },
 291             'add_ie': ['Dailymotion'],
 292         },
 293         # YouTube embed
 294         {
 295             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 296             'info_dict': {
 297                 'id': 'FXRb4ykk4S0',
 298                 'ext': 'mp4',
 299                 'title': 'The NBL Auction 2014',
 300                 'uploader': 'BADMINTON England',
 301                 'uploader_id': 'BADMINTONEvents',
 302                 'upload_date': '20140603',
 303                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 304             },
 305             'add_ie': ['Youtube'],
 306             'params': {
 307                 'skip_download': True,
 308             }
 309         },
 310         # MTVSercices embed
 311         {
 312             'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
 313             'md5': '35727f82f58c76d996fc188f9755b0d5',
 314             'info_dict': {
 315                 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
 316                 'ext': 'mp4',
 317                 'title': 'Review',
 318                 'description': 'Mario\'s life in the fast lane has never looked so good.',
 319             },
 320         },
 321         # YouTube embed via <data-embed-url="">
 322         {
 323             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 324             'info_dict': {
 325                 'id': '4vAffPZIT44',
 326                 'ext': 'mp4',
 327                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 328                 'uploader': 'Gameloft',
 329                 'uploader_id': 'gameloft',
 330                 'upload_date': '20140828',
 331                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 332             },
 333             'params': {
 334                 'skip_download': True,
 335             }
 336         },
 337         # Camtasia studio
 338         {
 339             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 340             'playlist': [{
 341                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 342                 'info_dict': {
 343                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 344                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 345                     'ext': 'flv',
 346                     'duration': 2235.90,
 347                 }
 348             }, {
 349                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 350                 'info_dict': {
 351                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 352                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 353                     'ext': 'flv',
 354                     'duration': 2235.93,
 355                 }
 356             }],
 357             'info_dict': {
 358                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 359             }
 360         },
 361         # Flowplayer
 362         {
 363             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 364             'md5': '9d65602bf31c6e20014319c7d07fba27',
 365             'info_dict': {
 366                 'id': '5123ea6d5e5a7',
 367                 'ext': 'mp4',
 368                 'age_limit': 18,
 369                 'uploader': 'www.handjobhub.com',
 370                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 371             }
 372         },
 373         # RSS feed
 374         {
 375             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 376             'info_dict': {
 377                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 378                 'title': 'Zero Punctuation',
 379                 'description': 're:.*groundbreaking video review series.*'
 380             },
 381             'playlist_mincount': 11,
 382         },
 383         # Multiple brightcove videos
 384         # https://github.com/rg3/youtube-dl/issues/2283
 385         {
 386             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 387             'info_dict': {
 388                 'id': 'always-never',
 389                 'title': 'Always / Never - The New Yorker',
 390             },
 391             'playlist_count': 3,
 392             'params': {
 393                 'extract_flat': False,
 394                 'skip_download': True,
 395             }
 396         },
 397         # MLB embed
 398         {
 399             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 400             'md5': '96f09a37e44da40dd083e12d9a683327',
 401             'info_dict': {
 402                 'id': '33322633',
 403                 'ext': 'mp4',
 404                 'title': 'Ump changes call to ball',
 405                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 406                 'duration': 48,
 407                 'timestamp': 1401537900,
 408                 'upload_date': '20140531',
 409                 'thumbnail': 're:^https?://.*\.jpg$',
 410             },
 411         },
 412         # Wistia embed
 413         {
 414             'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 415             'md5': '8788b683c777a5cf25621eaf286d0c23',
 416             'info_dict': {
 417                 'id': '1cfaf6b7ea',
 418                 'ext': 'mov',
 419                 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
 420                 'duration': 643.0,
 421                 'filesize': 182808282,
 422                 'uploader': 'education-portal.com',
 423             },
 424         },
 425         {
 426             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 427             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 428             'info_dict': {
 429                 'id': 'uxjb0lwrcz',
 430                 'ext': 'mp4',
 431                 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
 432                 'duration': 1715.0,
 433                 'uploader': 'thoughtworks.wistia.com',
 434             },
 435         },
 436         # Direct download with broken HEAD
 437         {
 438             'url': 'http://ai-radio.org:8000/radio.opus',
 439             'info_dict': {
 440                 'id': 'radio',
 441                 'ext': 'opus',
 442                 'title': 'radio',
 443             },
 444             'params': {
 445                 'skip_download': True,  # infinite live stream
 446             },
 447             'expected_warnings': [
 448                 r'501.*Not Implemented'
 449             ],
 450         },
 451         # Soundcloud embed
 452         {
 453             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 454             'info_dict': {
 455                 'id': '174391317',
 456                 'ext': 'mp3',
 457                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 458                 'uploader': 'Sophos Security',
 459                 'title': 'Chet Chat 171 - Oct 29, 2014',
 460                 'upload_date': '20141029',
 461             }
 462         },
 463         # Livestream embed
 464         {
 465             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 466             'info_dict': {
 467                 'id': '67864563',
 468                 'ext': 'flv',
 469                 'upload_date': '20141112',
 470                 'title': 'Rosetta #CometLanding webcast HL 10',
 471             }
 472         },
 473         # LazyYT
 474         {
 475             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
 476             'info_dict': {
 477                 'id': '1986',
 478                 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
 479             },
 480             'playlist_mincount': 2,
 481         },
 482         # Direct link with incorrect MIME type
 483         {
 484             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 485             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 486             'info_dict': {
 487                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 488                 'id': '5_Lennart_Poettering_-_Systemd',
 489                 'ext': 'webm',
 490                 'title': '5_Lennart_Poettering_-_Systemd',
 491                 'upload_date': '20141120',
 492             },
 493             'expected_warnings': [
 494                 'URL could be a direct video link, returning it as such.'
 495             ]
 496         },
 497         # Cinchcast embed
 498         {
 499             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
 500             'info_dict': {
 501                 'id': '7141703',
 502                 'ext': 'mp3',
 503                 'upload_date': '20141126',
 504                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
 505             }
 506         },
 507         # Cinerama player
 508         {
 509             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
 510             'info_dict': {
 511                 'id': '730m_DandD_1901_512k',
 512                 'ext': 'mp4',
 513                 'uploader': 'www.abc.net.au',
 514                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
 515             }
 516         },
 517         # embedded viddler video
 518         {
 519             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
 520             'info_dict': {
 521                 'id': '4d03aad9',
 522                 'ext': 'mp4',
 523                 'uploader': 'deadspin',
 524                 'title': 'WALL-TO-GORTAT',
 525                 'timestamp': 1422285291,
 526                 'upload_date': '20150126',
 527             },
 528             'add_ie': ['Viddler'],
 529         },
 530         # jwplayer YouTube
 531         {
 532             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
 533             'info_dict': {
 534                 'id': 'Mrj4DVp2zeA',
 535                 'ext': 'mp4',
 536                 'upload_date': '20150212',
 537                 'uploader': 'The National Archives UK',
 538                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
 539                 'uploader_id': 'NationalArchives08',
 540                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
 541             },
 542         },
 543         # rtl.nl embed
 544         {
 545             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
 546             'playlist_mincount': 5,
 547             'info_dict': {
 548                 'id': 'aanslagen-kopenhagen',
 549                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
 550             }
 551         },
 552         # Zapiks embed
 553         {
 554             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
 555             'info_dict': {
 556                 'id': '118046',
 557                 'ext': 'mp4',
 558                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
 559             }
 560         },
 561         # Kaltura embed
 562         {
 563             'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
 564             'info_dict': {
 565                 'id': '1_eergr3h1',
 566                 'ext': 'mp4',
 567                 'upload_date': '20150226',
 568                 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
 569                 'timestamp': int,
 570                 'title': 'John Carlson Postgame 2/25/15',
 571             },
 572         },
 573         # Eagle.Platform embed (generic URL)
 574         {
 575             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
 576             'info_dict': {
 577                 'id': '227304',
 578                 'ext': 'mp4',
 579                 'title': 'Навальный вышел на свободу',
 580                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
 581                 'thumbnail': 're:^https?://.*\.jpg$',
 582                 'duration': 87,
 583                 'view_count': int,
 584                 'age_limit': 0,
 585             },
 586         },
 587         # ClipYou (Eagle.Platform) embed (custom URL)
 588         {
 589             'url': 'http://muz-tv.ru/play/7129/',
 590             'info_dict': {
 591                 'id': '12820',
 592                 'ext': 'mp4',
 593                 'title': "'O Sole Mio",
 594                 'thumbnail': 're:^https?://.*\.jpg$',
 595                 'duration': 216,
 596                 'view_count': int,
 597             },
 598         },
 599         # Pladform embed
 600         {
 601             'url': 'http://muz-tv.ru/kinozal/view/7400/',
 602             'info_dict': {
 603                 'id': '100183293',
 604                 'ext': 'mp4',
 605                 'title': 'Тайны перевала Дятлова • Тайна перевала Дятлова 1 серия 2 часть',
 606                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
 607                 'thumbnail': 're:^https?://.*\.jpg$',
 608                 'duration': 694,
 609                 'age_limit': 0,
 610             },
 611         },
 612         # RSS feed with enclosure
 613         {
 614             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 615             'info_dict': {
 616                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 617                 'ext': 'm4v',
 618                 'upload_date': '20150228',
 619                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 620             }
 621         }
 622     ]
 623
 624     def report_following_redirect(self, new_url):
 625         """Report information extraction."""
 626         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
 627
 628     def _extract_rss(self, url, video_id, doc):
 629         playlist_title = doc.find('./channel/title').text
 630         playlist_desc_el = doc.find('./channel/description')
 631         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
 632
 633         entries = []
 634         for it in doc.findall('./channel/item'):
 635             next_url = xpath_text(it, 'link', fatal=False)
 636             if not next_url:
 637                 enclosure_nodes = it.findall('./enclosure')
 638                 for e in enclosure_nodes:
 639                     next_url = e.attrib.get('url')
 640                     if next_url:
 641                         break
 642
 643             if not next_url:
 644                 continue
 645
 646             entries.append({
 647                 '_type': 'url',
 648                 'url': next_url,
 649                 'title': it.find('title').text,
 650             })
 651
 652         return {
 653             '_type': 'playlist',
 654             'id': url,
 655             'title': playlist_title,
 656             'description': playlist_desc,
 657             'entries': entries,
 658         }
 659
 660     def _extract_camtasia(self, url, video_id, webpage):
 661         """ Returns None if no camtasia video can be found. """
 662
 663         camtasia_cfg = self._search_regex(
 664             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
 665             webpage, 'camtasia configuration file', default=None)
 666         if camtasia_cfg is None:
 667             return None
 668
 669         title = self._html_search_meta('DC.title', webpage, fatal=True)
 670
 671         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
 672         camtasia_cfg = self._download_xml(
 673             camtasia_url, video_id,
 674             note='Downloading camtasia configuration',
 675             errnote='Failed to download camtasia configuration')
 676         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
 677
 678         entries = []
 679         for n in fileset_node.getchildren():
 680             url_n = n.find('./uri')
 681             if url_n is None:
 682                 continue
 683
 684             entries.append({
 685                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
 686                 'title': '%s - %s' % (title, n.tag),
 687                 'url': compat_urlparse.urljoin(url, url_n.text),
 688                 'duration': float_or_none(n.find('./duration').text),
 689             })
 690
 691         return {
 692             '_type': 'playlist',
 693             'entries': entries,
 694             'title': title,
 695         }
 696
 697     def _real_extract(self, url):
 698         if url.startswith('//'):
 699             return {
 700                 '_type': 'url',
 701                 'url': self.http_scheme() + url,
 702             }
 703
 704         parsed_url = compat_urlparse.urlparse(url)
 705         if not parsed_url.scheme:
 706             default_search = self._downloader.params.get('default_search')
 707             if default_search is None:
 708                 default_search = 'fixup_error'
 709
 710             if default_search in ('auto', 'auto_warning', 'fixup_error'):
 711                 if '/' in url:
 712                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
 713                     return self.url_result('http://' + url)
 714                 elif default_search != 'fixup_error':
 715                     if default_search == 'auto_warning':
 716                         if re.match(r'^(?:url|URL)$', url):
 717                             raise ExtractorError(
 718                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
 719                                 expected=True)
 720                         else:
 721                             self._downloader.report_warning(
 722                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
 723                     return self.url_result('ytsearch:' + url)
 724
 725             if default_search in ('error', 'fixup_error'):
 726                 raise ExtractorError(
 727                     '%r is not a valid URL. '
 728                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
 729                     % (url, url), expected=True)
 730             else:
 731                 if ':' not in default_search:
 732                     default_search += ':'
 733                 return self.url_result(default_search + url)
 734
 735         url, smuggled_data = unsmuggle_url(url)
 736         force_videoid = None
 737         is_intentional = smuggled_data and smuggled_data.get('to_generic')
 738         if smuggled_data and 'force_videoid' in smuggled_data:
 739             force_videoid = smuggled_data['force_videoid']
 740             video_id = force_videoid
 741         else:
 742             video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
 743
 744         self.to_screen('%s: Requesting header' % video_id)
 745
 746         head_req = HEADRequest(url)
 747         head_response = self._request_webpage(
 748             head_req, video_id,
 749             note=False, errnote='Could not send HEAD request to %s' % url,
 750             fatal=False)
 751
 752         if head_response is not False:
 753             # Check for redirect
 754             new_url = head_response.geturl()
 755             if url != new_url:
 756                 self.report_following_redirect(new_url)
 757                 if force_videoid:
 758                     new_url = smuggle_url(
 759                         new_url, {'force_videoid': force_videoid})
 760                 return self.url_result(new_url)
 761
 762         full_response = None
 763         if head_response is False:
 764             full_response = self._request_webpage(url, video_id)
 765             head_response = full_response
 766
 767         # Check for direct link to a video
 768         content_type = head_response.headers.get('Content-Type', '')
 769         m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
 770         if m:
 771             upload_date = unified_strdate(
 772                 head_response.headers.get('Last-Modified'))
 773             return {
 774                 'id': video_id,
 775                 'title': os.path.splitext(url_basename(url))[0],
 776                 'direct': True,
 777                 'formats': [{
 778                     'format_id': m.group('format_id'),
 779                     'url': url,
 780                     'vcodec': 'none' if m.group('type') == 'audio' else None
 781                 }],
 782                 'upload_date': upload_date,
 783             }
 784
 785         if not self._downloader.params.get('test', False) and not is_intentional:
 786             self._downloader.report_warning('Falling back on generic information extractor.')
 787
 788         if not full_response:
 789             full_response = self._request_webpage(url, video_id)
 790
 791         # Maybe it's a direct link to a video?
 792         # Be careful not to download the whole thing!
 793         first_bytes = full_response.read(512)
 794         if not is_html(first_bytes):
 795             self._downloader.report_warning(
 796                 'URL could be a direct video link, returning it as such.')
 797             upload_date = unified_strdate(
 798                 head_response.headers.get('Last-Modified'))
 799             return {
 800                 'id': video_id,
 801                 'title': os.path.splitext(url_basename(url))[0],
 802                 'direct': True,
 803                 'url': url,
 804                 'upload_date': upload_date,
 805             }
 806
 807         webpage = self._webpage_read_content(
 808             full_response, url, video_id, prefix=first_bytes)
 809
 810         self.report_extraction(video_id)
 811
 812         # Is it an RSS feed?
 813         try:
 814             doc = parse_xml(webpage)
 815             if doc.tag == 'rss':
 816                 return self._extract_rss(url, video_id, doc)
 817         except compat_xml_parse_error:
 818             pass
 819
 820         # Is it a Camtasia project?
 821         camtasia_res = self._extract_camtasia(url, video_id, webpage)
 822         if camtasia_res is not None:
 823             return camtasia_res
 824
 825         # Sometimes embedded video player is hidden behind percent encoding
 826         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
 827         # Unescaping the whole page allows to handle those cases in a generic way
 828         webpage = compat_urllib_parse.unquote(webpage)
 829
 830         # it's tempting to parse this further, but you would
 831         # have to take into account all the variations like
 832         #   Video Title - Site Name
 833         #   Site Name | Video Title
 834         #   Video Title - Tagline | Site Name
 835         # and so on and so forth; it's just not practical
 836         video_title = self._html_search_regex(
 837             r'(?s)<title>(.*?)</title>', webpage, 'video title',
 838             default='video')
 839
 840         # Try to detect age limit automatically
 841         age_limit = self._rta_search(webpage)
 842         # And then there are the jokers who advertise that they use RTA,
 843         # but actually don't.
 844         AGE_LIMIT_MARKERS = [
 845             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
 846         ]
 847         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
 848             age_limit = 18
 849
 850         # video uploader is domain name
 851         video_uploader = self._search_regex(
 852             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
 853
 854         # Helper method
 855         def _playlist_from_matches(matches, getter=None, ie=None):
 856             urlrs = orderedSet(
 857                 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
 858                 for m in matches)
 859             return self.playlist_result(
 860                 urlrs, playlist_id=video_id, playlist_title=video_title)
 861
 862         # Look for BrightCove:
 863         bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
 864         if bc_urls:
 865             self.to_screen('Brightcove video detected.')
 866             entries = [{
 867                 '_type': 'url',
 868                 'url': smuggle_url(bc_url, {'Referer': url}),
 869                 'ie_key': 'Brightcove'
 870             } for bc_url in bc_urls]
 871
 872             return {
 873                 '_type': 'playlist',
 874                 'title': video_title,
 875                 'id': video_id,
 876                 'entries': entries,
 877             }
 878
 879         # Look for embedded rtl.nl player
 880         matches = re.findall(
 881             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+video_embed[^"]+)"',
 882             webpage)
 883         if matches:
 884             return _playlist_from_matches(matches, ie='RtlNl')
 885
 886         # Look for embedded (iframe) Vimeo player
 887         mobj = re.search(
 888             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
 889         if mobj:
 890             player_url = unescapeHTML(mobj.group('url'))
 891             surl = smuggle_url(player_url, {'Referer': url})
 892             return self.url_result(surl)
 893         # Look for embedded (swf embed) Vimeo player
 894         mobj = re.search(
 895             r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
 896         if mobj:
 897             return self.url_result(mobj.group(1))
 898
 899         # Look for embedded YouTube player
 900         matches = re.findall(r'''(?x)
 901             (?:
 902                 <iframe[^>]+?src=|
 903                 data-video-url=|
 904                 <embed[^>]+?src=|
 905                 embedSWF\(?:\s*|
 906                 new\s+SWFObject\(
 907             )
 908             (["\'])
 909                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
 910                 (?:embed|v|p)/.+?)
 911             \1''', webpage)
 912         if matches:
 913             return _playlist_from_matches(
 914                 matches, lambda m: unescapeHTML(m[1]))
 915
 916         # Look for lazyYT YouTube embed
 917         matches = re.findall(
 918             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
 919         if matches:
 920             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
 921
 922         # Look for embedded Dailymotion player
 923         matches = re.findall(
 924             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
 925         if matches:
 926             return _playlist_from_matches(
 927                 matches, lambda m: unescapeHTML(m[1]))
 928
 929         # Look for embedded Dailymotion playlist player (#3822)
 930         m = re.search(
 931             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
 932         if m:
 933             playlists = re.findall(
 934                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
 935             if playlists:
 936                 return _playlist_from_matches(
 937                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
 938
 939         # Look for embedded Wistia player
 940         match = re.search(
 941             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
 942         if match:
 943             embed_url = self._proto_relative_url(
 944                 unescapeHTML(match.group('url')))
 945             return {
 946                 '_type': 'url_transparent',
 947                 'url': embed_url,
 948                 'ie_key': 'Wistia',
 949                 'uploader': video_uploader,
 950                 'title': video_title,
 951                 'id': video_id,
 952             }
 953
 954         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
 955         if match:
 956             return {
 957                 '_type': 'url_transparent',
 958                 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
 959                 'ie_key': 'Wistia',
 960                 'uploader': video_uploader,
 961                 'title': video_title,
 962                 'id': match.group('id')
 963             }
 964
 965         # Look for embedded blip.tv player
 966         mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
 967         if mobj:
 968             return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
 969         mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
 970         if mobj:
 971             return self.url_result(mobj.group(1), 'BlipTV')
 972
 973         # Look for embedded condenast player
 974         matches = re.findall(
 975             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
 976             webpage)
 977         if matches:
 978             return {
 979                 '_type': 'playlist',
 980                 'entries': [{
 981                     '_type': 'url',
 982                     'ie_key': 'CondeNast',
 983                     'url': ma,
 984                 } for ma in matches],
 985                 'title': video_title,
 986                 'id': video_id,
 987             }
 988
 989         # Look for Bandcamp pages with custom domain
 990         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
 991         if mobj is not None:
 992             burl = unescapeHTML(mobj.group(1))
 993             # Don't set the extractor because it can be a track url or an album
 994             return self.url_result(burl)
 995
 996         # Look for embedded Vevo player
 997         mobj = re.search(
 998             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
 999         if mobj is not None:
1000             return self.url_result(mobj.group('url'))
1001
1002         # Look for embedded Viddler player
1003         mobj = re.search(
1004             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1005             webpage)
1006         if mobj is not None:
1007             return self.url_result(mobj.group('url'))
1008
1009         # Look for NYTimes player
1010         mobj = re.search(
1011             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1012             webpage)
1013         if mobj is not None:
1014             return self.url_result(mobj.group('url'))
1015
1016         # Look for Ooyala videos
1017         mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1018                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1019                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage))
1020         if mobj is not None:
1021             return OoyalaIE._build_url_result(mobj.group('ec'))
1022
1023         # Look for multiple Ooyala embeds on SBN network websites
1024         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1025         if mobj is not None:
1026             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1027             if embeds:
1028                 return _playlist_from_matches(
1029                     embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
1030
1031         # Look for Aparat videos
1032         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1033         if mobj is not None:
1034             return self.url_result(mobj.group(1), 'Aparat')
1035
1036         # Look for MPORA videos
1037         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1038         if mobj is not None:
1039             return self.url_result(mobj.group(1), 'Mpora')
1040
1041         # Look for embedded NovaMov-based player
1042         mobj = re.search(
1043             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1044                     (?P<url>http://(?:(?:embed|www)\.)?
1045                         (?:novamov\.com|
1046                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
1047                            videoweed\.(?:es|com)|
1048                            movshare\.(?:net|sx|ag)|
1049                            divxstage\.(?:eu|net|ch|co|at|ag))
1050                         /embed\.php.+?)\1''', webpage)
1051         if mobj is not None:
1052             return self.url_result(mobj.group('url'))
1053
1054         # Look for embedded Facebook player
1055         mobj = re.search(
1056             r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1057         if mobj is not None:
1058             return self.url_result(mobj.group('url'), 'Facebook')
1059
1060         # Look for embedded VK player
1061         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1062         if mobj is not None:
1063             return self.url_result(mobj.group('url'), 'VK')
1064
1065         # Look for embedded ivi player
1066         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1067         if mobj is not None:
1068             return self.url_result(mobj.group('url'), 'Ivi')
1069
1070         # Look for embedded Huffington Post player
1071         mobj = re.search(
1072             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1073         if mobj is not None:
1074             return self.url_result(mobj.group('url'), 'HuffPost')
1075
1076         # Look for embed.ly
1077         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1078         if mobj is not None:
1079             return self.url_result(mobj.group('url'))
1080         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1081         if mobj is not None:
1082             return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
1083
1084         # Look for funnyordie embed
1085         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1086         if matches:
1087             return _playlist_from_matches(
1088                 matches, getter=unescapeHTML, ie='FunnyOrDie')
1089
1090         # Look for BBC iPlayer embed
1091         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1092         if matches:
1093             return _playlist_from_matches(matches, ie='BBCCoUk')
1094
1095         # Look for embedded RUTV player
1096         rutv_url = RUTVIE._extract_url(webpage)
1097         if rutv_url:
1098             return self.url_result(rutv_url, 'RUTV')
1099
1100         # Look for embedded TED player
1101         mobj = re.search(
1102             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1103         if mobj is not None:
1104             return self.url_result(mobj.group('url'), 'TED')
1105
1106         # Look for embedded Ustream videos
1107         mobj = re.search(
1108             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1109         if mobj is not None:
1110             return self.url_result(mobj.group('url'), 'Ustream')
1111
1112         # Look for embedded arte.tv player
1113         mobj = re.search(
1114             r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1115             webpage)
1116         if mobj is not None:
1117             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1118
1119         # Look for embedded smotri.com player
1120         smotri_url = SmotriIE._extract_url(webpage)
1121         if smotri_url:
1122             return self.url_result(smotri_url, 'Smotri')
1123
1124         # Look for embeded soundcloud player
1125         mobj = re.search(
1126             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1127             webpage)
1128         if mobj is not None:
1129             url = unescapeHTML(mobj.group('url'))
1130             return self.url_result(url)
1131
1132         # Look for embedded vulture.com player
1133         mobj = re.search(
1134             r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1135             webpage)
1136         if mobj is not None:
1137             url = unescapeHTML(mobj.group('url'))
1138             return self.url_result(url, ie='Vulture')
1139
1140         # Look for embedded mtvservices player
1141         mobj = re.search(
1142             r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1143             webpage)
1144         if mobj is not None:
1145             url = unescapeHTML(mobj.group('url'))
1146             return self.url_result(url, ie='MTVServicesEmbedded')
1147
1148         # Look for embedded yahoo player
1149         mobj = re.search(
1150             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1151             webpage)
1152         if mobj is not None:
1153             return self.url_result(mobj.group('url'), 'Yahoo')
1154
1155         # Look for embedded sbs.com.au player
1156         mobj = re.search(
1157             r'''(?x)
1158             (?:
1159                 <meta\s+property="og:video"\s+content=|
1160                 <iframe[^>]+?src=
1161             )
1162             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1163             webpage)
1164         if mobj is not None:
1165             return self.url_result(mobj.group('url'), 'SBS')
1166
1167         # Look for embedded Cinchcast player
1168         mobj = re.search(
1169             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1170             webpage)
1171         if mobj is not None:
1172             return self.url_result(mobj.group('url'), 'Cinchcast')
1173
1174         mobj = re.search(
1175             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1176             webpage)
1177         if mobj is not None:
1178             return self.url_result(mobj.group('url'), 'MLB')
1179
1180         mobj = re.search(
1181             r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1182             webpage)
1183         if mobj is not None:
1184             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1185
1186         mobj = re.search(
1187             r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1188             webpage)
1189         if mobj is not None:
1190             return self.url_result(mobj.group('url'), 'Livestream')
1191
1192         # Look for Zapiks embed
1193         mobj = re.search(
1194             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1195         if mobj is not None:
1196             return self.url_result(mobj.group('url'), 'Zapiks')
1197
1198         # Look for Kaltura embeds
1199         mobj = re.search(
1200             r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
1201         if mobj is not None:
1202             return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
1203
1204         # Look for Eagle.Platform embeds
1205         mobj = re.search(
1206             r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1207         if mobj is not None:
1208             return self.url_result(mobj.group('url'), 'EaglePlatform')
1209
1210         # Look for ClipYou (uses Eagle.Platform) embeds
1211         mobj = re.search(
1212             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1213         if mobj is not None:
1214             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1215
1216         # Look for Pladform embeds
1217         mobj = re.search(
1218             r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
1219         if mobj is not None:
1220             return self.url_result(mobj.group('url'), 'Pladform')
1221
1222         def check_video(vurl):
1223             if YoutubeIE.suitable(vurl):
1224                 return True
1225             vpath = compat_urlparse.urlparse(vurl).path
1226             vext = determine_ext(vpath)
1227             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1228
1229         def filter_video(urls):
1230             return list(filter(check_video, urls))
1231
1232         # Start with something easy: JW Player in SWFObject
1233         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1234         if not found:
1235             # Look for gorilla-vid style embedding
1236             found = filter_video(re.findall(r'''(?sx)
1237                 (?:
1238                     jw_plugins|
1239                     JWPlayerOptions|
1240                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1241                 )
1242                 .*?
1243                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1244         if not found:
1245             # Broaden the search a little bit
1246             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1247         if not found:
1248             # Broaden the findall a little bit: JWPlayer JS loader
1249             found = filter_video(re.findall(
1250                 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1251         if not found:
1252             # Flow player
1253             found = filter_video(re.findall(r'''(?xs)
1254                 flowplayer\("[^"]+",\s*
1255                     \{[^}]+?\}\s*,
1256                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1257                         ["']?url["']?\s*:\s*["']([^"']+)["']
1258             ''', webpage))
1259         if not found:
1260             # Cinerama player
1261             found = re.findall(
1262                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1263         if not found:
1264             # Try to find twitter cards info
1265             found = filter_video(re.findall(
1266                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1267         if not found:
1268             # We look for Open Graph info:
1269             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1270             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1271             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1272             if m_video_type is not None:
1273                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1274         if not found:
1275             # HTML5 video
1276             found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1277         if not found:
1278             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1279             found = re.search(
1280                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1281                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1282                 webpage)
1283             if not found:
1284                 # Look also in Refresh HTTP header
1285                 refresh_header = head_response.headers.get('Refresh')
1286                 if refresh_header:
1287                     found = re.search(REDIRECT_REGEX, refresh_header)
1288             if found:
1289                 new_url = found.group(1)
1290                 self.report_following_redirect(new_url)
1291                 return {
1292                     '_type': 'url',
1293                     'url': new_url,
1294                 }
1295         if not found:
1296             raise UnsupportedError(url)
1297
1298         entries = []
1299         for video_url in found:
1300             video_url = compat_urlparse.urljoin(url, video_url)
1301             video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1302
1303             # Sometimes, jwplayer extraction will result in a YouTube URL
1304             if YoutubeIE.suitable(video_url):
1305                 entries.append(self.url_result(video_url, 'Youtube'))
1306                 continue
1307
1308             # here's a fun little line of code for you:
1309             video_id = os.path.splitext(video_id)[0]
1310
1311             entries.append({
1312                 'id': video_id,
1313                 'url': video_url,
1314                 'uploader': video_uploader,
1315                 'title': video_title,
1316                 'age_limit': age_limit,
1317             })
1318
1319         if len(entries) == 1:
1320             return entries[0]
1321         else:
1322             for num, e in enumerate(entries, start=1):
1323                 # 'url' results don't have a title
1324                 if e.get('title') is not None:
1325                     e['title'] = '%s (%d)' % (e['title'], num)
1326             return {
1327                 '_type': 'playlist',
1328                 'entries': entries,
1329             }