_ 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_urllib_parse_unquote,
  13     compat_urllib_request,
  14     compat_urlparse,
  15     compat_xml_parse_error,
  16 )
  17 from ..utils import (
  18     determine_ext,
  19     ExtractorError,
  20     float_or_none,
  21     HEADRequest,
  22     is_html,
  23     orderedSet,
  24     parse_xml,
  25     smuggle_url,
  26     unescapeHTML,
  27     unified_strdate,
  28     unsmuggle_url,
  29     UnsupportedError,
  30     url_basename,
  31     xpath_text,
  32 )
  33 from .brightcove import BrightcoveIE
  34 from .nbc import NBCSportsVPlayerIE
  35 from .ooyala import OoyalaIE
  36 from .rutv import RUTVIE
  37 from .tvc import TVCIE
  38 from .sportbox import SportBoxEmbedIE
  39 from .smotri import SmotriIE
  40 from .condenast import CondeNastIE
  41 from .udn import UDNEmbedIE
  42 from .senateisvp import SenateISVPIE
  43 from .bliptv import BlipTVIE
  44 from .svt import SVTIE
  45 from .pornhub import PornHubIE
  46 from .xhamster import XHamsterEmbedIE
  47 from .vimeo import VimeoIE
  48 from .dailymotion import DailymotionCloudIE
  49
  50
  51 class GenericIE(InfoExtractor):
  52     IE_DESC = 'Generic downloader that works on some sites'
  53     _VALID_URL = r'.*'
  54     IE_NAME = 'generic'
  55     _TESTS = [
  56         # Direct link to a video
  57         {
  58             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  59             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  60             'info_dict': {
  61                 'id': 'trailer',
  62                 'ext': 'mp4',
  63                 'title': 'trailer',
  64                 'upload_date': '20100513',
  65             }
  66         },
  67         # Direct link to media delivered compressed (until Accept-Encoding is *)
  68         {
  69             'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
  70             'md5': '128c42e68b13950268b648275386fc74',
  71             'info_dict': {
  72                 'id': 'FictionJunction-Parallel_Hearts',
  73                 'ext': 'flac',
  74                 'title': 'FictionJunction-Parallel_Hearts',
  75                 'upload_date': '20140522',
  76             },
  77             'expected_warnings': [
  78                 'URL could be a direct video link, returning it as such.'
  79             ]
  80         },
  81         # Direct download with broken HEAD
  82         {
  83             'url': 'http://ai-radio.org:8000/radio.opus',
  84             'info_dict': {
  85                 'id': 'radio',
  86                 'ext': 'opus',
  87                 'title': 'radio',
  88             },
  89             'params': {
  90                 'skip_download': True,  # infinite live stream
  91             },
  92             'expected_warnings': [
  93                 r'501.*Not Implemented'
  94             ],
  95         },
  96         # Direct link with incorrect MIME type
  97         {
  98             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
  99             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 100             'info_dict': {
 101                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 102                 'id': '5_Lennart_Poettering_-_Systemd',
 103                 'ext': 'webm',
 104                 'title': '5_Lennart_Poettering_-_Systemd',
 105                 'upload_date': '20141120',
 106             },
 107             'expected_warnings': [
 108                 'URL could be a direct video link, returning it as such.'
 109             ]
 110         },
 111         # RSS feed
 112         {
 113             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 114             'info_dict': {
 115                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 116                 'title': 'Zero Punctuation',
 117                 'description': 're:.*groundbreaking video review series.*'
 118             },
 119             'playlist_mincount': 11,
 120         },
 121         # RSS feed with enclosure
 122         {
 123             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 124             'info_dict': {
 125                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 126                 'ext': 'm4v',
 127                 'upload_date': '20150228',
 128                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 129             }
 130         },
 131         # google redirect
 132         {
 133             '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',
 134             'info_dict': {
 135                 'id': 'cmQHVoWB5FY',
 136                 'ext': 'mp4',
 137                 'upload_date': '20130224',
 138                 'uploader_id': 'TheVerge',
 139                 'description': 're:^Chris Ziegler takes a look at the\.*',
 140                 'uploader': 'The Verge',
 141                 'title': 'First Firefox OS phones side-by-side',
 142             },
 143             'params': {
 144                 'skip_download': False,
 145             }
 146         },
 147         {
 148             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
 149             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
 150             'info_dict': {
 151                 'id': '13601338388002',
 152                 'ext': 'mp4',
 153                 'uploader': 'www.hodiho.fr',
 154                 'title': 'R\u00e9gis plante sa Jeep',
 155             }
 156         },
 157         # bandcamp page with custom domain
 158         {
 159             'add_ie': ['Bandcamp'],
 160             'url': 'http://bronyrock.com/track/the-pony-mash',
 161             'info_dict': {
 162                 'id': '3235767654',
 163                 'ext': 'mp3',
 164                 'title': 'The Pony Mash',
 165                 'uploader': 'M_Pallante',
 166             },
 167             'skip': 'There is a limit of 200 free downloads / month for the test song',
 168         },
 169         # embedded brightcove video
 170         # it also tests brightcove videos that need to set the 'Referer' in the
 171         # http requests
 172         {
 173             'add_ie': ['Brightcove'],
 174             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
 175             'info_dict': {
 176                 'id': '2765128793001',
 177                 'ext': 'mp4',
 178                 'title': 'Le cours de bourse : l’analyse technique',
 179                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
 180                 'uploader': 'BFM BUSINESS',
 181             },
 182             'params': {
 183                 'skip_download': True,
 184             },
 185         },
 186         {
 187             # https://github.com/rg3/youtube-dl/issues/2253
 188             'url': 'http://bcove.me/i6nfkrc3',
 189             'md5': '0ba9446db037002366bab3b3eb30c88c',
 190             'info_dict': {
 191                 'id': '3101154703001',
 192                 'ext': 'mp4',
 193                 'title': 'Still no power',
 194                 'uploader': 'thestar.com',
 195                 '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.',
 196             },
 197             'add_ie': ['Brightcove'],
 198         },
 199         {
 200             'url': 'http://www.championat.com/video/football/v/87/87499.html',
 201             'md5': 'fb973ecf6e4a78a67453647444222983',
 202             'info_dict': {
 203                 'id': '3414141473001',
 204                 'ext': 'mp4',
 205                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 206                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 207                 'uploader': 'Championat',
 208             },
 209         },
 210         {
 211             # https://github.com/rg3/youtube-dl/issues/3541
 212             'add_ie': ['Brightcove'],
 213             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 214             'info_dict': {
 215                 'id': '3866516442001',
 216                 'ext': 'mp4',
 217                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 218                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 219                 'uploader': 'SBS Broadcasting',
 220             },
 221             'skip': 'Restricted to Netherlands',
 222             'params': {
 223                 'skip_download': True,  # m3u8 download
 224             },
 225         },
 226         # ooyala video
 227         {
 228             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 229             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 230             'info_dict': {
 231                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 232                 'ext': 'mp4',
 233                 'title': '2cc213299525360.mov',  # that's what we get
 234             },
 235             'add_ie': ['Ooyala'],
 236         },
 237         # multiple ooyala embeds on SBN network websites
 238         {
 239             'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 240             'info_dict': {
 241                 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 242                 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
 243             },
 244             'playlist_mincount': 3,
 245             'params': {
 246                 'skip_download': True,
 247             },
 248             'add_ie': ['Ooyala'],
 249         },
 250         # embed.ly video
 251         {
 252             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 253             'info_dict': {
 254                 'id': '9ODmcdjQcHQ',
 255                 'ext': 'mp4',
 256                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 257                 'upload_date': '20140225',
 258                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 259                 'uploader': 'Tested',
 260                 'uploader_id': 'testedcom',
 261             },
 262             # No need to test YoutubeIE here
 263             'params': {
 264                 'skip_download': True,
 265             },
 266         },
 267         # funnyordie embed
 268         {
 269             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 270             'info_dict': {
 271                 'id': '18e820ec3f',
 272                 'ext': 'mp4',
 273                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 274                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 275             },
 276         },
 277         # BBC iPlayer embeds
 278         {
 279             'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
 280             'info_dict': {
 281                 'title': 'BBC - Blogs -  Adam Curtis - BUGGER',
 282             },
 283             'playlist_mincount': 18,
 284         },
 285         # RUTV embed
 286         {
 287             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 288             'info_dict': {
 289                 'id': '776940',
 290                 'ext': 'mp4',
 291                 'title': 'Охотское море стало целиком российским',
 292                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 293             },
 294             'params': {
 295                 # m3u8 download
 296                 'skip_download': True,
 297             },
 298         },
 299         # TVC embed
 300         {
 301             'url': 'http://sch1298sz.mskobr.ru/dou_edu/karamel_ki/filial_galleries/video/iframe_src_http_tvc_ru_video_iframe_id_55304_isplay_false_acc_video_id_channel_brand_id_11_show_episodes_episode_id_32307_frameb/',
 302             'info_dict': {
 303                 'id': '55304',
 304                 'ext': 'mp4',
 305                 'title': 'Дошкольное воспитание',
 306             },
 307         },
 308         # SportBox embed
 309         {
 310             'url': 'http://www.vestifinance.ru/articles/25753',
 311             'info_dict': {
 312                 'id': '25753',
 313                 'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
 314             },
 315             'playlist': [{
 316                 'info_dict': {
 317                     'id': '370908',
 318                     'title': 'Госзаказ. День 3',
 319                     'ext': 'mp4',
 320                 }
 321             }, {
 322                 'info_dict': {
 323                     'id': '370905',
 324                     'title': 'Госзаказ. День 2',
 325                     'ext': 'mp4',
 326                 }
 327             }, {
 328                 'info_dict': {
 329                     'id': '370902',
 330                     'title': 'Госзаказ. День 1',
 331                     'ext': 'mp4',
 332                 }
 333             }],
 334             'params': {
 335                 # m3u8 download
 336                 'skip_download': True,
 337             },
 338         },
 339         # Embedded TED video
 340         {
 341             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 342             'md5': '65fdff94098e4a607385a60c5177c638',
 343             'info_dict': {
 344                 'id': '1969',
 345                 'ext': 'mp4',
 346                 'title': 'Hidden miracles of the natural world',
 347                 'uploader': 'Louie Schwartzberg',
 348                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 349             }
 350         },
 351         # Embeded Ustream video
 352         {
 353             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 354             'md5': '27b99cdb639c9b12a79bca876a073417',
 355             'info_dict': {
 356                 'id': '45734260',
 357                 'ext': 'flv',
 358                 'uploader': 'AU SPA:  The NSA and Privacy',
 359                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 360             }
 361         },
 362         # nowvideo embed hidden behind percent encoding
 363         {
 364             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 365             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 366             'info_dict': {
 367                 'id': '06e53103ca9aa',
 368                 'ext': 'flv',
 369                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 370                 'description': 'No description',
 371             },
 372         },
 373         # arte embed
 374         {
 375             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 376             'md5': '7653032cbb25bf6c80d80f217055fa43',
 377             'info_dict': {
 378                 'id': '048195-004_PLUS7-F',
 379                 'ext': 'flv',
 380                 'title': 'X:enius',
 381                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 382                 'upload_date': '20140320',
 383             },
 384             'params': {
 385                 'skip_download': 'Requires rtmpdump'
 386             }
 387         },
 388         # Condé Nast embed
 389         {
 390             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 391             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 392             'info_dict': {
 393                 'id': '53501be369702d3275860000',
 394                 'ext': 'mp4',
 395                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 396             }
 397         },
 398         # Dailymotion embed
 399         {
 400             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 401             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 402             'info_dict': {
 403                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 404                 'ext': 'mp4',
 405                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 406                 'uploader': 'Spi0n',
 407             },
 408             'add_ie': ['Dailymotion'],
 409         },
 410         # YouTube embed
 411         {
 412             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 413             'info_dict': {
 414                 'id': 'FXRb4ykk4S0',
 415                 'ext': 'mp4',
 416                 'title': 'The NBL Auction 2014',
 417                 'uploader': 'BADMINTON England',
 418                 'uploader_id': 'BADMINTONEvents',
 419                 'upload_date': '20140603',
 420                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 421             },
 422             'add_ie': ['Youtube'],
 423             'params': {
 424                 'skip_download': True,
 425             }
 426         },
 427         # MTVSercices embed
 428         {
 429             'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
 430             'md5': '35727f82f58c76d996fc188f9755b0d5',
 431             'info_dict': {
 432                 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
 433                 'ext': 'mp4',
 434                 'title': 'Review',
 435                 'description': 'Mario\'s life in the fast lane has never looked so good.',
 436             },
 437         },
 438         # YouTube embed via <data-embed-url="">
 439         {
 440             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 441             'info_dict': {
 442                 'id': '4vAffPZIT44',
 443                 'ext': 'mp4',
 444                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 445                 'uploader': 'Gameloft',
 446                 'uploader_id': 'gameloft',
 447                 'upload_date': '20140828',
 448                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 449             },
 450             'params': {
 451                 'skip_download': True,
 452             }
 453         },
 454         # Camtasia studio
 455         {
 456             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 457             'playlist': [{
 458                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 459                 'info_dict': {
 460                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 461                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 462                     'ext': 'flv',
 463                     'duration': 2235.90,
 464                 }
 465             }, {
 466                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 467                 'info_dict': {
 468                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 469                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 470                     'ext': 'flv',
 471                     'duration': 2235.93,
 472                 }
 473             }],
 474             'info_dict': {
 475                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 476             }
 477         },
 478         # Flowplayer
 479         {
 480             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 481             'md5': '9d65602bf31c6e20014319c7d07fba27',
 482             'info_dict': {
 483                 'id': '5123ea6d5e5a7',
 484                 'ext': 'mp4',
 485                 'age_limit': 18,
 486                 'uploader': 'www.handjobhub.com',
 487                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 488             }
 489         },
 490         # Multiple brightcove videos
 491         # https://github.com/rg3/youtube-dl/issues/2283
 492         {
 493             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 494             'info_dict': {
 495                 'id': 'always-never',
 496                 'title': 'Always / Never - The New Yorker',
 497             },
 498             'playlist_count': 3,
 499             'params': {
 500                 'extract_flat': False,
 501                 'skip_download': True,
 502             }
 503         },
 504         # MLB embed
 505         {
 506             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 507             'md5': '96f09a37e44da40dd083e12d9a683327',
 508             'info_dict': {
 509                 'id': '33322633',
 510                 'ext': 'mp4',
 511                 'title': 'Ump changes call to ball',
 512                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 513                 'duration': 48,
 514                 'timestamp': 1401537900,
 515                 'upload_date': '20140531',
 516                 'thumbnail': 're:^https?://.*\.jpg$',
 517             },
 518         },
 519         # Wistia embed
 520         {
 521             'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 522             'md5': '8788b683c777a5cf25621eaf286d0c23',
 523             'info_dict': {
 524                 'id': '1cfaf6b7ea',
 525                 'ext': 'mov',
 526                 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
 527                 'duration': 643.0,
 528                 'filesize': 182808282,
 529                 'uploader': 'education-portal.com',
 530             },
 531         },
 532         {
 533             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 534             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 535             'info_dict': {
 536                 'id': 'uxjb0lwrcz',
 537                 'ext': 'mp4',
 538                 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
 539                 'duration': 1715.0,
 540                 'uploader': 'thoughtworks.wistia.com',
 541             },
 542         },
 543         # Soundcloud embed
 544         {
 545             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 546             'info_dict': {
 547                 'id': '174391317',
 548                 'ext': 'mp3',
 549                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 550                 'uploader': 'Sophos Security',
 551                 'title': 'Chet Chat 171 - Oct 29, 2014',
 552                 'upload_date': '20141029',
 553             }
 554         },
 555         # Livestream embed
 556         {
 557             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 558             'info_dict': {
 559                 'id': '67864563',
 560                 'ext': 'flv',
 561                 'upload_date': '20141112',
 562                 'title': 'Rosetta #CometLanding webcast HL 10',
 563             }
 564         },
 565         # LazyYT
 566         {
 567             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
 568             'info_dict': {
 569                 'id': '1986',
 570                 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
 571             },
 572             'playlist_mincount': 2,
 573         },
 574         # Cinchcast embed
 575         {
 576             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
 577             'info_dict': {
 578                 'id': '7141703',
 579                 'ext': 'mp3',
 580                 'upload_date': '20141126',
 581                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
 582             }
 583         },
 584         # Cinerama player
 585         {
 586             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
 587             'info_dict': {
 588                 'id': '730m_DandD_1901_512k',
 589                 'ext': 'mp4',
 590                 'uploader': 'www.abc.net.au',
 591                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
 592             }
 593         },
 594         # embedded viddler video
 595         {
 596             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
 597             'info_dict': {
 598                 'id': '4d03aad9',
 599                 'ext': 'mp4',
 600                 'uploader': 'deadspin',
 601                 'title': 'WALL-TO-GORTAT',
 602                 'timestamp': 1422285291,
 603                 'upload_date': '20150126',
 604             },
 605             'add_ie': ['Viddler'],
 606         },
 607         # Libsyn embed
 608         {
 609             'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
 610             'info_dict': {
 611                 'id': '3377616',
 612                 'ext': 'mp3',
 613                 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
 614                 'description': 'md5:601cb790edd05908957dae8aaa866465',
 615                 'upload_date': '20150220',
 616             },
 617         },
 618         # jwplayer YouTube
 619         {
 620             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
 621             'info_dict': {
 622                 'id': 'Mrj4DVp2zeA',
 623                 'ext': 'mp4',
 624                 'upload_date': '20150212',
 625                 'uploader': 'The National Archives UK',
 626                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
 627                 'uploader_id': 'NationalArchives08',
 628                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
 629             },
 630         },
 631         # rtl.nl embed
 632         {
 633             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
 634             'playlist_mincount': 5,
 635             'info_dict': {
 636                 'id': 'aanslagen-kopenhagen',
 637                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
 638             }
 639         },
 640         # Zapiks embed
 641         {
 642             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
 643             'info_dict': {
 644                 'id': '118046',
 645                 'ext': 'mp4',
 646                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
 647             }
 648         },
 649         # Kaltura embed
 650         {
 651             'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
 652             'info_dict': {
 653                 'id': '1_eergr3h1',
 654                 'ext': 'mp4',
 655                 'upload_date': '20150226',
 656                 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
 657                 'timestamp': int,
 658                 'title': 'John Carlson Postgame 2/25/15',
 659             },
 660         },
 661         # Eagle.Platform embed (generic URL)
 662         {
 663             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
 664             'info_dict': {
 665                 'id': '227304',
 666                 'ext': 'mp4',
 667                 'title': 'Навальный вышел на свободу',
 668                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
 669                 'thumbnail': 're:^https?://.*\.jpg$',
 670                 'duration': 87,
 671                 'view_count': int,
 672                 'age_limit': 0,
 673             },
 674         },
 675         # ClipYou (Eagle.Platform) embed (custom URL)
 676         {
 677             'url': 'http://muz-tv.ru/play/7129/',
 678             'info_dict': {
 679                 'id': '12820',
 680                 'ext': 'mp4',
 681                 'title': "'O Sole Mio",
 682                 'thumbnail': 're:^https?://.*\.jpg$',
 683                 'duration': 216,
 684                 'view_count': int,
 685             },
 686         },
 687         # Pladform embed
 688         {
 689             'url': 'http://muz-tv.ru/kinozal/view/7400/',
 690             'info_dict': {
 691                 'id': '100183293',
 692                 'ext': 'mp4',
 693                 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
 694                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
 695                 'thumbnail': 're:^https?://.*\.jpg$',
 696                 'duration': 694,
 697                 'age_limit': 0,
 698             },
 699         },
 700         # Playwire embed
 701         {
 702             'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
 703             'info_dict': {
 704                 'id': '3519514',
 705                 'ext': 'mp4',
 706                 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
 707                 'thumbnail': 're:^https?://.*\.png$',
 708                 'duration': 45.115,
 709             },
 710         },
 711         # 5min embed
 712         {
 713             'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
 714             'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
 715             'info_dict': {
 716                 'id': '518726732',
 717                 'ext': 'mp4',
 718                 'title': 'Facebook Creates "On This Day" | Crunch Report',
 719             },
 720         },
 721         # SVT embed
 722         {
 723             'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 724             'info_dict': {
 725                 'id': '2900353',
 726                 'ext': 'flv',
 727                 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
 728                 'duration': 27,
 729                 'age_limit': 0,
 730             },
 731         },
 732         # Crooks and Liars embed
 733         {
 734             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
 735             'info_dict': {
 736                 'id': '8RUoRhRi',
 737                 'ext': 'mp4',
 738                 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
 739                 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
 740                 'timestamp': 1428207000,
 741                 'upload_date': '20150405',
 742                 'uploader': 'Heather',
 743             },
 744         },
 745         # Crooks and Liars external embed
 746         {
 747             'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
 748             'info_dict': {
 749                 'id': 'MTE3MjUtMzQ2MzA',
 750                 'ext': 'mp4',
 751                 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
 752                 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
 753                 'timestamp': 1265032391,
 754                 'upload_date': '20100201',
 755                 'uploader': 'Heather',
 756             },
 757         },
 758         # NBC Sports vplayer embed
 759         {
 760             'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
 761             'info_dict': {
 762                 'id': 'ln7x1qSThw4k',
 763                 'ext': 'flv',
 764                 'title': "PFT Live: New leader in the 'new-look' defense",
 765                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
 766             },
 767         },
 768         # UDN embed
 769         {
 770             'url': 'http://www.udn.com/news/story/7314/822787',
 771             'md5': 'fd2060e988c326991037b9aff9df21a6',
 772             'info_dict': {
 773                 'id': '300346',
 774                 'ext': 'mp4',
 775                 'title': '中一中男師變性 全校師生力挺',
 776                 'thumbnail': 're:^https?://.*\.jpg$',
 777             }
 778         },
 779         # Ooyala embed
 780         {
 781             'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
 782             'info_dict': {
 783                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
 784                 'ext': 'mp4',
 785                 'description': 'VIDEO: Index/Match versus VLOOKUP.',
 786                 'title': 'This is what separates the Excel masters from the wannabes',
 787             },
 788             'params': {
 789                 # m3u8 downloads
 790                 'skip_download': True,
 791             }
 792         },
 793         # Contains a SMIL manifest
 794         {
 795             'url': 'http://www.telewebion.com/fa/1263668/%D9%82%D8%B1%D8%B9%D9%87%E2%80%8C%DA%A9%D8%B4%DB%8C-%D9%84%DB%8C%DA%AF-%D9%82%D9%87%D8%B1%D9%85%D8%A7%D9%86%D8%A7%D9%86-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7/%2B-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84.html',
 796             'info_dict': {
 797                 'id': 'file',
 798                 'ext': 'flv',
 799                 'title': '+ Football: Lottery Champions League Europe',
 800                 'uploader': 'www.telewebion.com',
 801             },
 802             'params': {
 803                 # rtmpe downloads
 804                 'skip_download': True,
 805             }
 806         },
 807         # Brightcove URL in single quotes
 808         {
 809             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
 810             'md5': '4ae374f1f8b91c889c4b9203c8c752af',
 811             'info_dict': {
 812                 'id': '4255764656001',
 813                 'ext': 'mp4',
 814                 'title': 'SN Presents: Russell Martin, World Citizen',
 815                 'description': 'To understand why he was the Toronto Blue Jays’ top off-season priority is to appreciate his background and upbringing in Montreal, where he first developed his baseball skills. Written and narrated by Stephen Brunt.',
 816                 'uploader': 'Rogers Sportsnet',
 817             },
 818         },
 819         # Dailymotion Cloud video
 820         {
 821             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
 822             'md5': '49444254273501a64675a7e68c502681',
 823             'info_dict': {
 824                 'id': '5585de919473990de4bee11b',
 825                 'ext': 'mp4',
 826                 'title': 'Le débat',
 827                 'thumbnail': 're:^https?://.*\.jpe?g$',
 828             }
 829         }
 830     ]
 831
 832     def report_following_redirect(self, new_url):
 833         """Report information extraction."""
 834         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
 835
 836     def _extract_rss(self, url, video_id, doc):
 837         playlist_title = doc.find('./channel/title').text
 838         playlist_desc_el = doc.find('./channel/description')
 839         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
 840
 841         entries = []
 842         for it in doc.findall('./channel/item'):
 843             next_url = xpath_text(it, 'link', fatal=False)
 844             if not next_url:
 845                 enclosure_nodes = it.findall('./enclosure')
 846                 for e in enclosure_nodes:
 847                     next_url = e.attrib.get('url')
 848                     if next_url:
 849                         break
 850
 851             if not next_url:
 852                 continue
 853
 854             entries.append({
 855                 '_type': 'url',
 856                 'url': next_url,
 857                 'title': it.find('title').text,
 858             })
 859
 860         return {
 861             '_type': 'playlist',
 862             'id': url,
 863             'title': playlist_title,
 864             'description': playlist_desc,
 865             'entries': entries,
 866         }
 867
 868     def _extract_camtasia(self, url, video_id, webpage):
 869         """ Returns None if no camtasia video can be found. """
 870
 871         camtasia_cfg = self._search_regex(
 872             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
 873             webpage, 'camtasia configuration file', default=None)
 874         if camtasia_cfg is None:
 875             return None
 876
 877         title = self._html_search_meta('DC.title', webpage, fatal=True)
 878
 879         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
 880         camtasia_cfg = self._download_xml(
 881             camtasia_url, video_id,
 882             note='Downloading camtasia configuration',
 883             errnote='Failed to download camtasia configuration')
 884         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
 885
 886         entries = []
 887         for n in fileset_node.getchildren():
 888             url_n = n.find('./uri')
 889             if url_n is None:
 890                 continue
 891
 892             entries.append({
 893                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
 894                 'title': '%s - %s' % (title, n.tag),
 895                 'url': compat_urlparse.urljoin(url, url_n.text),
 896                 'duration': float_or_none(n.find('./duration').text),
 897             })
 898
 899         return {
 900             '_type': 'playlist',
 901             'entries': entries,
 902             'title': title,
 903         }
 904
 905     def _real_extract(self, url):
 906         if url.startswith('//'):
 907             return {
 908                 '_type': 'url',
 909                 'url': self.http_scheme() + url,
 910             }
 911
 912         parsed_url = compat_urlparse.urlparse(url)
 913         if not parsed_url.scheme:
 914             default_search = self._downloader.params.get('default_search')
 915             if default_search is None:
 916                 default_search = 'fixup_error'
 917
 918             if default_search in ('auto', 'auto_warning', 'fixup_error'):
 919                 if '/' in url:
 920                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
 921                     return self.url_result('http://' + url)
 922                 elif default_search != 'fixup_error':
 923                     if default_search == 'auto_warning':
 924                         if re.match(r'^(?:url|URL)$', url):
 925                             raise ExtractorError(
 926                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
 927                                 expected=True)
 928                         else:
 929                             self._downloader.report_warning(
 930                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
 931                     return self.url_result('ytsearch:' + url)
 932
 933             if default_search in ('error', 'fixup_error'):
 934                 raise ExtractorError(
 935                     '%r is not a valid URL. '
 936                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
 937                     % (url, url), expected=True)
 938             else:
 939                 if ':' not in default_search:
 940                     default_search += ':'
 941                 return self.url_result(default_search + url)
 942
 943         url, smuggled_data = unsmuggle_url(url)
 944         force_videoid = None
 945         is_intentional = smuggled_data and smuggled_data.get('to_generic')
 946         if smuggled_data and 'force_videoid' in smuggled_data:
 947             force_videoid = smuggled_data['force_videoid']
 948             video_id = force_videoid
 949         else:
 950             video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
 951
 952         self.to_screen('%s: Requesting header' % video_id)
 953
 954         head_req = HEADRequest(url)
 955         head_response = self._request_webpage(
 956             head_req, video_id,
 957             note=False, errnote='Could not send HEAD request to %s' % url,
 958             fatal=False)
 959
 960         if head_response is not False:
 961             # Check for redirect
 962             new_url = head_response.geturl()
 963             if url != new_url:
 964                 self.report_following_redirect(new_url)
 965                 if force_videoid:
 966                     new_url = smuggle_url(
 967                         new_url, {'force_videoid': force_videoid})
 968                 return self.url_result(new_url)
 969
 970         full_response = None
 971         if head_response is False:
 972             request = compat_urllib_request.Request(url)
 973             request.add_header('Accept-Encoding', '*')
 974             full_response = self._request_webpage(request, video_id)
 975             head_response = full_response
 976
 977         # Check for direct link to a video
 978         content_type = head_response.headers.get('Content-Type', '')
 979         m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
 980         if m:
 981             upload_date = unified_strdate(
 982                 head_response.headers.get('Last-Modified'))
 983             return {
 984                 'id': video_id,
 985                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
 986                 'direct': True,
 987                 'formats': [{
 988                     'format_id': m.group('format_id'),
 989                     'url': url,
 990                     'vcodec': 'none' if m.group('type') == 'audio' else None
 991                 }],
 992                 'upload_date': upload_date,
 993             }
 994
 995         if not self._downloader.params.get('test', False) and not is_intentional:
 996             self._downloader.report_warning('Falling back on generic information extractor.')
 997
 998         if not full_response:
 999             request = compat_urllib_request.Request(url)
1000             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
1001             # making it impossible to download only chunk of the file (yet we need only 512kB to
1002             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
1003             # that will always result in downloading the whole file that is not desirable.
1004             # Therefore for extraction pass we have to override Accept-Encoding to any in order
1005             # to accept raw bytes and being able to download only a chunk.
1006             # It may probably better to solve this by checking Content-Type for application/octet-stream
1007             # after HEAD request finishes, but not sure if we can rely on this.
1008             request.add_header('Accept-Encoding', '*')
1009             full_response = self._request_webpage(request, video_id)
1010
1011         # Maybe it's a direct link to a video?
1012         # Be careful not to download the whole thing!
1013         first_bytes = full_response.read(512)
1014         if not is_html(first_bytes):
1015             self._downloader.report_warning(
1016                 'URL could be a direct video link, returning it as such.')
1017             upload_date = unified_strdate(
1018                 head_response.headers.get('Last-Modified'))
1019             return {
1020                 'id': video_id,
1021                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1022                 'direct': True,
1023                 'url': url,
1024                 'upload_date': upload_date,
1025             }
1026
1027         webpage = self._webpage_read_content(
1028             full_response, url, video_id, prefix=first_bytes)
1029
1030         self.report_extraction(video_id)
1031
1032         # Is it an RSS feed?
1033         try:
1034             doc = parse_xml(webpage)
1035             if doc.tag == 'rss':
1036                 return self._extract_rss(url, video_id, doc)
1037         except compat_xml_parse_error:
1038             pass
1039
1040         # Is it a Camtasia project?
1041         camtasia_res = self._extract_camtasia(url, video_id, webpage)
1042         if camtasia_res is not None:
1043             return camtasia_res
1044
1045         # Sometimes embedded video player is hidden behind percent encoding
1046         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1047         # Unescaping the whole page allows to handle those cases in a generic way
1048         webpage = compat_urllib_parse.unquote(webpage)
1049
1050         # it's tempting to parse this further, but you would
1051         # have to take into account all the variations like
1052         #   Video Title - Site Name
1053         #   Site Name | Video Title
1054         #   Video Title - Tagline | Site Name
1055         # and so on and so forth; it's just not practical
1056         video_title = self._html_search_regex(
1057             r'(?s)<title>(.*?)</title>', webpage, 'video title',
1058             default='video')
1059
1060         # Try to detect age limit automatically
1061         age_limit = self._rta_search(webpage)
1062         # And then there are the jokers who advertise that they use RTA,
1063         # but actually don't.
1064         AGE_LIMIT_MARKERS = [
1065             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1066         ]
1067         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1068             age_limit = 18
1069
1070         # video uploader is domain name
1071         video_uploader = self._search_regex(
1072             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1073
1074         # Helper method
1075         def _playlist_from_matches(matches, getter=None, ie=None):
1076             urlrs = orderedSet(
1077                 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1078                 for m in matches)
1079             return self.playlist_result(
1080                 urlrs, playlist_id=video_id, playlist_title=video_title)
1081
1082         # Look for BrightCove:
1083         bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
1084         if bc_urls:
1085             self.to_screen('Brightcove video detected.')
1086             entries = [{
1087                 '_type': 'url',
1088                 'url': smuggle_url(bc_url, {'Referer': url}),
1089                 'ie_key': 'Brightcove'
1090             } for bc_url in bc_urls]
1091
1092             return {
1093                 '_type': 'playlist',
1094                 'title': video_title,
1095                 'id': video_id,
1096                 'entries': entries,
1097             }
1098
1099         # Look for embedded rtl.nl player
1100         matches = re.findall(
1101             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1102             webpage)
1103         if matches:
1104             return _playlist_from_matches(matches, ie='RtlNl')
1105
1106         vimeo_url = VimeoIE._extract_vimeo_url(url, webpage)
1107         if vimeo_url is not None:
1108             return self.url_result(vimeo_url)
1109
1110         # Look for embedded YouTube player
1111         matches = re.findall(r'''(?x)
1112             (?:
1113                 <iframe[^>]+?src=|
1114                 data-video-url=|
1115                 <embed[^>]+?src=|
1116                 embedSWF\(?:\s*|
1117                 new\s+SWFObject\(
1118             )
1119             (["\'])
1120                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1121                 (?:embed|v|p)/.+?)
1122             \1''', webpage)
1123         if matches:
1124             return _playlist_from_matches(
1125                 matches, lambda m: unescapeHTML(m[1]))
1126
1127         # Look for lazyYT YouTube embed
1128         matches = re.findall(
1129             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1130         if matches:
1131             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1132
1133         # Look for embedded Dailymotion player
1134         matches = re.findall(
1135             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
1136         if matches:
1137             return _playlist_from_matches(
1138                 matches, lambda m: unescapeHTML(m[1]))
1139
1140         # Look for embedded Dailymotion playlist player (#3822)
1141         m = re.search(
1142             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1143         if m:
1144             playlists = re.findall(
1145                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1146             if playlists:
1147                 return _playlist_from_matches(
1148                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1149
1150         # Look for embedded Wistia player
1151         match = re.search(
1152             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1153         if match:
1154             embed_url = self._proto_relative_url(
1155                 unescapeHTML(match.group('url')))
1156             return {
1157                 '_type': 'url_transparent',
1158                 'url': embed_url,
1159                 'ie_key': 'Wistia',
1160                 'uploader': video_uploader,
1161                 'title': video_title,
1162                 'id': video_id,
1163             }
1164
1165         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1166         if match:
1167             return {
1168                 '_type': 'url_transparent',
1169                 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1170                 'ie_key': 'Wistia',
1171                 'uploader': video_uploader,
1172                 'title': video_title,
1173                 'id': match.group('id')
1174             }
1175
1176         # Look for embedded blip.tv player
1177         bliptv_url = BlipTVIE._extract_url(webpage)
1178         if bliptv_url:
1179             return self.url_result(bliptv_url, 'BlipTV')
1180
1181         # Look for SVT player
1182         svt_url = SVTIE._extract_url(webpage)
1183         if svt_url:
1184             return self.url_result(svt_url, 'SVT')
1185
1186         # Look for embedded condenast player
1187         matches = re.findall(
1188             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1189             webpage)
1190         if matches:
1191             return {
1192                 '_type': 'playlist',
1193                 'entries': [{
1194                     '_type': 'url',
1195                     'ie_key': 'CondeNast',
1196                     'url': ma,
1197                 } for ma in matches],
1198                 'title': video_title,
1199                 'id': video_id,
1200             }
1201
1202         # Look for Bandcamp pages with custom domain
1203         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1204         if mobj is not None:
1205             burl = unescapeHTML(mobj.group(1))
1206             # Don't set the extractor because it can be a track url or an album
1207             return self.url_result(burl)
1208
1209         # Look for embedded Vevo player
1210         mobj = re.search(
1211             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1212         if mobj is not None:
1213             return self.url_result(mobj.group('url'))
1214
1215         # Look for embedded Viddler player
1216         mobj = re.search(
1217             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1218             webpage)
1219         if mobj is not None:
1220             return self.url_result(mobj.group('url'))
1221
1222         # Look for NYTimes player
1223         mobj = re.search(
1224             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1225             webpage)
1226         if mobj is not None:
1227             return self.url_result(mobj.group('url'))
1228
1229         # Look for Libsyn player
1230         mobj = re.search(
1231             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1232         if mobj is not None:
1233             return self.url_result(mobj.group('url'))
1234
1235         # Look for Ooyala videos
1236         mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1237                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1238                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1239                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1240         if mobj is not None:
1241             return OoyalaIE._build_url_result(mobj.group('ec'))
1242
1243         # Look for multiple Ooyala embeds on SBN network websites
1244         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1245         if mobj is not None:
1246             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1247             if embeds:
1248                 return _playlist_from_matches(
1249                     embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
1250
1251         # Look for Aparat videos
1252         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1253         if mobj is not None:
1254             return self.url_result(mobj.group(1), 'Aparat')
1255
1256         # Look for MPORA videos
1257         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1258         if mobj is not None:
1259             return self.url_result(mobj.group(1), 'Mpora')
1260
1261         # Look for embedded NovaMov-based player
1262         mobj = re.search(
1263             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1264                     (?P<url>http://(?:(?:embed|www)\.)?
1265                         (?:novamov\.com|
1266                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
1267                            videoweed\.(?:es|com)|
1268                            movshare\.(?:net|sx|ag)|
1269                            divxstage\.(?:eu|net|ch|co|at|ag))
1270                         /embed\.php.+?)\1''', webpage)
1271         if mobj is not None:
1272             return self.url_result(mobj.group('url'))
1273
1274         # Look for embedded Facebook player
1275         mobj = re.search(
1276             r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1277         if mobj is not None:
1278             return self.url_result(mobj.group('url'), 'Facebook')
1279
1280         # Look for embedded VK player
1281         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1282         if mobj is not None:
1283             return self.url_result(mobj.group('url'), 'VK')
1284
1285         # Look for embedded ivi player
1286         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1287         if mobj is not None:
1288             return self.url_result(mobj.group('url'), 'Ivi')
1289
1290         # Look for embedded Huffington Post player
1291         mobj = re.search(
1292             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1293         if mobj is not None:
1294             return self.url_result(mobj.group('url'), 'HuffPost')
1295
1296         # Look for embed.ly
1297         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1298         if mobj is not None:
1299             return self.url_result(mobj.group('url'))
1300         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1301         if mobj is not None:
1302             return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
1303
1304         # Look for funnyordie embed
1305         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1306         if matches:
1307             return _playlist_from_matches(
1308                 matches, getter=unescapeHTML, ie='FunnyOrDie')
1309
1310         # Look for BBC iPlayer embed
1311         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1312         if matches:
1313             return _playlist_from_matches(matches, ie='BBCCoUk')
1314
1315         # Look for embedded RUTV player
1316         rutv_url = RUTVIE._extract_url(webpage)
1317         if rutv_url:
1318             return self.url_result(rutv_url, 'RUTV')
1319
1320         # Look for embedded TVC player
1321         tvc_url = TVCIE._extract_url(webpage)
1322         if tvc_url:
1323             return self.url_result(tvc_url, 'TVC')
1324
1325         # Look for embedded SportBox player
1326         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1327         if sportbox_urls:
1328             return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1329
1330         # Look for embedded PornHub player
1331         pornhub_url = PornHubIE._extract_url(webpage)
1332         if pornhub_url:
1333             return self.url_result(pornhub_url, 'PornHub')
1334
1335         # Look for embedded XHamster player
1336         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
1337         if xhamster_urls:
1338             return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
1339
1340         # Look for embedded Tvigle player
1341         mobj = re.search(
1342             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
1343         if mobj is not None:
1344             return self.url_result(mobj.group('url'), 'Tvigle')
1345
1346         # Look for embedded TED player
1347         mobj = re.search(
1348             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1349         if mobj is not None:
1350             return self.url_result(mobj.group('url'), 'TED')
1351
1352         # Look for embedded Ustream videos
1353         mobj = re.search(
1354             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1355         if mobj is not None:
1356             return self.url_result(mobj.group('url'), 'Ustream')
1357
1358         # Look for embedded arte.tv player
1359         mobj = re.search(
1360             r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1361             webpage)
1362         if mobj is not None:
1363             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1364
1365         # Look for embedded smotri.com player
1366         smotri_url = SmotriIE._extract_url(webpage)
1367         if smotri_url:
1368             return self.url_result(smotri_url, 'Smotri')
1369
1370         # Look for embeded soundcloud player
1371         mobj = re.search(
1372             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1373             webpage)
1374         if mobj is not None:
1375             url = unescapeHTML(mobj.group('url'))
1376             return self.url_result(url)
1377
1378         # Look for embedded vulture.com player
1379         mobj = re.search(
1380             r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1381             webpage)
1382         if mobj is not None:
1383             url = unescapeHTML(mobj.group('url'))
1384             return self.url_result(url, ie='Vulture')
1385
1386         # Look for embedded mtvservices player
1387         mobj = re.search(
1388             r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1389             webpage)
1390         if mobj is not None:
1391             url = unescapeHTML(mobj.group('url'))
1392             return self.url_result(url, ie='MTVServicesEmbedded')
1393
1394         # Look for embedded yahoo player
1395         mobj = re.search(
1396             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1397             webpage)
1398         if mobj is not None:
1399             return self.url_result(mobj.group('url'), 'Yahoo')
1400
1401         # Look for embedded sbs.com.au player
1402         mobj = re.search(
1403             r'''(?x)
1404             (?:
1405                 <meta\s+property="og:video"\s+content=|
1406                 <iframe[^>]+?src=
1407             )
1408             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1409             webpage)
1410         if mobj is not None:
1411             return self.url_result(mobj.group('url'), 'SBS')
1412
1413         # Look for embedded Cinchcast player
1414         mobj = re.search(
1415             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1416             webpage)
1417         if mobj is not None:
1418             return self.url_result(mobj.group('url'), 'Cinchcast')
1419
1420         mobj = re.search(
1421             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1422             webpage)
1423         if not mobj:
1424             mobj = re.search(
1425                 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
1426                 webpage)
1427         if mobj is not None:
1428             return self.url_result(mobj.group('url'), 'MLB')
1429
1430         mobj = re.search(
1431             r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1432             webpage)
1433         if mobj is not None:
1434             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1435
1436         mobj = re.search(
1437             r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1438             webpage)
1439         if mobj is not None:
1440             return self.url_result(mobj.group('url'), 'Livestream')
1441
1442         # Look for Zapiks embed
1443         mobj = re.search(
1444             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1445         if mobj is not None:
1446             return self.url_result(mobj.group('url'), 'Zapiks')
1447
1448         # Look for Kaltura embeds
1449         mobj = re.search(
1450             r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
1451         if mobj is not None:
1452             return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
1453
1454         # Look for Eagle.Platform embeds
1455         mobj = re.search(
1456             r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1457         if mobj is not None:
1458             return self.url_result(mobj.group('url'), 'EaglePlatform')
1459
1460         # Look for ClipYou (uses Eagle.Platform) embeds
1461         mobj = re.search(
1462             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1463         if mobj is not None:
1464             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1465
1466         # Look for Pladform embeds
1467         mobj = re.search(
1468             r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
1469         if mobj is not None:
1470             return self.url_result(mobj.group('url'), 'Pladform')
1471
1472         # Look for Playwire embeds
1473         mobj = re.search(
1474             r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
1475         if mobj is not None:
1476             return self.url_result(mobj.group('url'))
1477
1478         # Look for 5min embeds
1479         mobj = re.search(
1480             r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1481         if mobj is not None:
1482             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1483
1484         # Look for Crooks and Liars embeds
1485         mobj = re.search(
1486             r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1487         if mobj is not None:
1488             return self.url_result(mobj.group('url'))
1489
1490         # Look for NBC Sports VPlayer embeds
1491         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1492         if nbc_sports_url:
1493             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1494
1495         # Look for UDN embeds
1496         mobj = re.search(
1497             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage)
1498         if mobj is not None:
1499             return self.url_result(
1500                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1501
1502         # Look for Senate ISVP iframe
1503         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
1504         if senate_isvp_url:
1505             return self.url_result(senate_isvp_url, 'SenateISVP')
1506
1507         # Look for Dailymotion Cloud videos
1508         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
1509         if dmcloud_url:
1510             return self.url_result(dmcloud_url, 'DailymotionCloud')
1511
1512         def check_video(vurl):
1513             if YoutubeIE.suitable(vurl):
1514                 return True
1515             vpath = compat_urlparse.urlparse(vurl).path
1516             vext = determine_ext(vpath)
1517             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1518
1519         def filter_video(urls):
1520             return list(filter(check_video, urls))
1521
1522         # Start with something easy: JW Player in SWFObject
1523         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1524         if not found:
1525             # Look for gorilla-vid style embedding
1526             found = filter_video(re.findall(r'''(?sx)
1527                 (?:
1528                     jw_plugins|
1529                     JWPlayerOptions|
1530                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1531                 )
1532                 .*?
1533                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1534         if not found:
1535             # Broaden the search a little bit
1536             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1537         if not found:
1538             # Broaden the findall a little bit: JWPlayer JS loader
1539             found = filter_video(re.findall(
1540                 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1541         if not found:
1542             # Flow player
1543             found = filter_video(re.findall(r'''(?xs)
1544                 flowplayer\("[^"]+",\s*
1545                     \{[^}]+?\}\s*,
1546                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1547                         ["']?url["']?\s*:\s*["']([^"']+)["']
1548             ''', webpage))
1549         if not found:
1550             # Cinerama player
1551             found = re.findall(
1552                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1553         if not found:
1554             # Try to find twitter cards info
1555             found = filter_video(re.findall(
1556                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1557         if not found:
1558             # We look for Open Graph info:
1559             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1560             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1561             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1562             if m_video_type is not None:
1563                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1564         if not found:
1565             # HTML5 video
1566             found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1567         if not found:
1568             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1569             found = re.search(
1570                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1571                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1572                 webpage)
1573             if not found:
1574                 # Look also in Refresh HTTP header
1575                 refresh_header = head_response.headers.get('Refresh')
1576                 if refresh_header:
1577                     found = re.search(REDIRECT_REGEX, refresh_header)
1578             if found:
1579                 new_url = compat_urlparse.urljoin(url, found.group(1))
1580                 self.report_following_redirect(new_url)
1581                 return {
1582                     '_type': 'url',
1583                     'url': new_url,
1584                 }
1585         if not found:
1586             raise UnsupportedError(url)
1587
1588         entries = []
1589         for video_url in found:
1590             video_url = compat_urlparse.urljoin(url, video_url)
1591             video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1592
1593             # Sometimes, jwplayer extraction will result in a YouTube URL
1594             if YoutubeIE.suitable(video_url):
1595                 entries.append(self.url_result(video_url, 'Youtube'))
1596                 continue
1597
1598             # here's a fun little line of code for you:
1599             video_id = os.path.splitext(video_id)[0]
1600
1601             if determine_ext(video_url) == 'smil':
1602                 entries.append({
1603                     'id': video_id,
1604                     'formats': self._extract_smil_formats(video_url, video_id),
1605                     'uploader': video_uploader,
1606                     'title': video_title,
1607                     'age_limit': age_limit,
1608                 })
1609             else:
1610                 entries.append({
1611                     'id': video_id,
1612                     'url': video_url,
1613                     'uploader': video_uploader,
1614                     'title': video_title,
1615                     'age_limit': age_limit,
1616                 })
1617
1618         if len(entries) == 1:
1619             return entries[0]
1620         else:
1621             for num, e in enumerate(entries, start=1):
1622                 # 'url' results don't have a title
1623                 if e.get('title') is not None:
1624                     e['title'] = '%s (%d)' % (e['title'], num)
1625             return {
1626                 '_type': 'playlist',
1627                 'entries': entries,
1628             }