_ 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 import sys
   8
   9 from .common import InfoExtractor
  10 from .youtube import YoutubeIE
  11 from ..compat import (
  12     compat_etree_fromstring,
  13     compat_urllib_parse_unquote,
  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     sanitized_Request,
  25     smuggle_url,
  26     unescapeHTML,
  27     unified_strdate,
  28     unsmuggle_url,
  29     UnsupportedError,
  30     url_basename,
  31     xpath_text,
  32 )
  33 from .brightcove import (
  34     BrightcoveLegacyIE,
  35     BrightcoveNewIE,
  36 )
  37 from .nbc import NBCSportsVPlayerIE
  38 from .ooyala import OoyalaIE
  39 from .rutv import RUTVIE
  40 from .tvc import TVCIE
  41 from .sportbox import SportBoxEmbedIE
  42 from .smotri import SmotriIE
  43 from .myvi import MyviIE
  44 from .condenast import CondeNastIE
  45 from .udn import UDNEmbedIE
  46 from .senateisvp import SenateISVPIE
  47 from .bliptv import BlipTVIE
  48 from .svt import SVTIE
  49 from .pornhub import PornHubIE
  50 from .xhamster import XHamsterEmbedIE
  51 from .vimeo import VimeoIE
  52 from .dailymotion import DailymotionCloudIE
  53 from .onionstudios import OnionStudiosIE
  54 from .snagfilms import SnagFilmsEmbedIE
  55 from .screenwavemedia import ScreenwaveMediaIE
  56 from .mtv import MTVServicesEmbeddedIE
  57
  58
  59 class GenericIE(InfoExtractor):
  60     IE_DESC = 'Generic downloader that works on some sites'
  61     _VALID_URL = r'.*'
  62     IE_NAME = 'generic'
  63     _TESTS = [
  64         # Direct link to a video
  65         {
  66             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  67             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  68             'info_dict': {
  69                 'id': 'trailer',
  70                 'ext': 'mp4',
  71                 'title': 'trailer',
  72                 'upload_date': '20100513',
  73             }
  74         },
  75         # Direct link to media delivered compressed (until Accept-Encoding is *)
  76         {
  77             'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
  78             'md5': '128c42e68b13950268b648275386fc74',
  79             'info_dict': {
  80                 'id': 'FictionJunction-Parallel_Hearts',
  81                 'ext': 'flac',
  82                 'title': 'FictionJunction-Parallel_Hearts',
  83                 'upload_date': '20140522',
  84             },
  85             'expected_warnings': [
  86                 'URL could be a direct video link, returning it as such.'
  87             ]
  88         },
  89         # Direct download with broken HEAD
  90         {
  91             'url': 'http://ai-radio.org:8000/radio.opus',
  92             'info_dict': {
  93                 'id': 'radio',
  94                 'ext': 'opus',
  95                 'title': 'radio',
  96             },
  97             'params': {
  98                 'skip_download': True,  # infinite live stream
  99             },
 100             'expected_warnings': [
 101                 r'501.*Not Implemented'
 102             ],
 103         },
 104         # Direct link with incorrect MIME type
 105         {
 106             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 107             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 108             'info_dict': {
 109                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 110                 'id': '5_Lennart_Poettering_-_Systemd',
 111                 'ext': 'webm',
 112                 'title': '5_Lennart_Poettering_-_Systemd',
 113                 'upload_date': '20141120',
 114             },
 115             'expected_warnings': [
 116                 'URL could be a direct video link, returning it as such.'
 117             ]
 118         },
 119         # RSS feed
 120         {
 121             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 122             'info_dict': {
 123                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 124                 'title': 'Zero Punctuation',
 125                 'description': 're:.*groundbreaking video review series.*'
 126             },
 127             'playlist_mincount': 11,
 128         },
 129         # RSS feed with enclosure
 130         {
 131             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 132             'info_dict': {
 133                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 134                 'ext': 'm4v',
 135                 'upload_date': '20150228',
 136                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 137             }
 138         },
 139         # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
 140         {
 141             'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
 142             'info_dict': {
 143                 'id': 'smil',
 144                 'ext': 'mp4',
 145                 'title': 'Automatics, robotics and biocybernetics',
 146                 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
 147                 'upload_date': '20130627',
 148                 'formats': 'mincount:16',
 149                 'subtitles': 'mincount:1',
 150             },
 151             'params': {
 152                 'force_generic_extractor': True,
 153                 'skip_download': True,
 154             },
 155         },
 156         # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html
 157         {
 158             'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil',
 159             'info_dict': {
 160                 'id': 'hds',
 161                 'ext': 'flv',
 162                 'title': 'hds',
 163                 'formats': 'mincount:1',
 164             },
 165             'params': {
 166                 'skip_download': True,
 167             },
 168         },
 169         # SMIL from https://www.restudy.dk/video/play/id/1637
 170         {
 171             'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml',
 172             'info_dict': {
 173                 'id': 'video_1637',
 174                 'ext': 'flv',
 175                 'title': 'video_1637',
 176                 'formats': 'mincount:3',
 177             },
 178             'params': {
 179                 'skip_download': True,
 180             },
 181         },
 182         # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm
 183         {
 184             'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil',
 185             'info_dict': {
 186                 'id': 'smil-service',
 187                 'ext': 'flv',
 188                 'title': 'smil-service',
 189                 'formats': 'mincount:1',
 190             },
 191             'params': {
 192                 'skip_download': True,
 193             },
 194         },
 195         # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370
 196         {
 197             'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil',
 198             'info_dict': {
 199                 'id': '4719370',
 200                 'ext': 'mp4',
 201                 'title': '571de1fd-47bc-48db-abf9-238872a58d1f',
 202                 'formats': 'mincount:3',
 203             },
 204             'params': {
 205                 'skip_download': True,
 206             },
 207         },
 208         # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html
 209         {
 210             'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf',
 211             'info_dict': {
 212                 'id': 'mZlp2ctYIUEB',
 213                 'ext': 'mp4',
 214                 'title': 'Tikibad ontruimd wegens brand',
 215                 'description': 'md5:05ca046ff47b931f9b04855015e163a4',
 216                 'thumbnail': 're:^https?://.*\.jpg$',
 217                 'duration': 33,
 218             },
 219             'params': {
 220                 'skip_download': True,
 221             },
 222         },
 223         # google redirect
 224         {
 225             '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',
 226             'info_dict': {
 227                 'id': 'cmQHVoWB5FY',
 228                 'ext': 'mp4',
 229                 'upload_date': '20130224',
 230                 'uploader_id': 'TheVerge',
 231                 'description': 're:^Chris Ziegler takes a look at the\.*',
 232                 'uploader': 'The Verge',
 233                 'title': 'First Firefox OS phones side-by-side',
 234             },
 235             'params': {
 236                 'skip_download': False,
 237             }
 238         },
 239         {
 240             # redirect in Refresh HTTP header
 241             'url': 'https://www.facebook.com/l.php?u=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DpO8h3EaFRdo&h=TAQHsoToz&enc=AZN16h-b6o4Zq9pZkCCdOLNKMN96BbGMNtcFwHSaazus4JHT_MFYkAA-WARTX2kvsCIdlAIyHZjl6d33ILIJU7Jzwk_K3mcenAXoAzBNoZDI_Q7EXGDJnIhrGkLXo_LJ_pAa2Jzbx17UHMd3jAs--6j2zaeto5w9RTn8T_1kKg3fdC5WPX9Dbb18vzH7YFX0eSJmoa6SP114rvlkw6pkS1-T&s=1',
 242             'info_dict': {
 243                 'id': 'pO8h3EaFRdo',
 244                 'ext': 'mp4',
 245                 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
 246                 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5',
 247                 'upload_date': '20150917',
 248                 'uploader_id': 'brtvofficial',
 249                 'uploader': 'Boiler Room',
 250             },
 251             'params': {
 252                 'skip_download': False,
 253             },
 254         },
 255         {
 256             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
 257             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
 258             'info_dict': {
 259                 'id': '13601338388002',
 260                 'ext': 'mp4',
 261                 'uploader': 'www.hodiho.fr',
 262                 'title': 'R\u00e9gis plante sa Jeep',
 263             }
 264         },
 265         # bandcamp page with custom domain
 266         {
 267             'add_ie': ['Bandcamp'],
 268             'url': 'http://bronyrock.com/track/the-pony-mash',
 269             'info_dict': {
 270                 'id': '3235767654',
 271                 'ext': 'mp3',
 272                 'title': 'The Pony Mash',
 273                 'uploader': 'M_Pallante',
 274             },
 275             'skip': 'There is a limit of 200 free downloads / month for the test song',
 276         },
 277         # embedded brightcove video
 278         # it also tests brightcove videos that need to set the 'Referer' in the
 279         # http requests
 280         {
 281             'add_ie': ['BrightcoveLegacy'],
 282             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
 283             'info_dict': {
 284                 'id': '2765128793001',
 285                 'ext': 'mp4',
 286                 'title': 'Le cours de bourse : l’analyse technique',
 287                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
 288                 'uploader': 'BFM BUSINESS',
 289             },
 290             'params': {
 291                 'skip_download': True,
 292             },
 293         },
 294         {
 295             # https://github.com/rg3/youtube-dl/issues/2253
 296             'url': 'http://bcove.me/i6nfkrc3',
 297             'md5': '0ba9446db037002366bab3b3eb30c88c',
 298             'info_dict': {
 299                 'id': '3101154703001',
 300                 'ext': 'mp4',
 301                 'title': 'Still no power',
 302                 'uploader': 'thestar.com',
 303                 '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.',
 304             },
 305             'add_ie': ['BrightcoveLegacy'],
 306         },
 307         {
 308             'url': 'http://www.championat.com/video/football/v/87/87499.html',
 309             'md5': 'fb973ecf6e4a78a67453647444222983',
 310             'info_dict': {
 311                 'id': '3414141473001',
 312                 'ext': 'mp4',
 313                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 314                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 315                 'uploader': 'Championat',
 316             },
 317         },
 318         {
 319             # https://github.com/rg3/youtube-dl/issues/3541
 320             'add_ie': ['BrightcoveLegacy'],
 321             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 322             'info_dict': {
 323                 'id': '3866516442001',
 324                 'ext': 'mp4',
 325                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 326                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 327                 'uploader': 'SBS Broadcasting',
 328             },
 329             'skip': 'Restricted to Netherlands',
 330             'params': {
 331                 'skip_download': True,  # m3u8 download
 332             },
 333         },
 334         # ooyala video
 335         {
 336             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 337             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 338             'info_dict': {
 339                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 340                 'ext': 'mp4',
 341                 'title': '2cc213299525360.mov',  # that's what we get
 342                 'duration': 238231,
 343             },
 344             'add_ie': ['Ooyala'],
 345         },
 346         {
 347             # ooyala video embedded with http://player.ooyala.com/iframe.js
 348             'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/',
 349             'info_dict': {
 350                 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB',
 351                 'ext': 'mp4',
 352                 'title': '"Steve Jobs: Man in the Machine" trailer',
 353                 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
 354                 'duration': 135427,
 355             },
 356             'params': {
 357                 'skip_download': True,
 358             },
 359         },
 360         # multiple ooyala embeds on SBN network websites
 361         {
 362             'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 363             'info_dict': {
 364                 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 365                 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
 366             },
 367             'playlist_mincount': 3,
 368             'params': {
 369                 'skip_download': True,
 370             },
 371             'add_ie': ['Ooyala'],
 372         },
 373         # embed.ly video
 374         {
 375             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 376             'info_dict': {
 377                 'id': '9ODmcdjQcHQ',
 378                 'ext': 'mp4',
 379                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 380                 'upload_date': '20140225',
 381                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 382                 'uploader': 'Tested',
 383                 'uploader_id': 'testedcom',
 384             },
 385             # No need to test YoutubeIE here
 386             'params': {
 387                 'skip_download': True,
 388             },
 389         },
 390         # funnyordie embed
 391         {
 392             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 393             'info_dict': {
 394                 'id': '18e820ec3f',
 395                 'ext': 'mp4',
 396                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 397                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 398             },
 399         },
 400         # RUTV embed
 401         {
 402             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 403             'info_dict': {
 404                 'id': '776940',
 405                 'ext': 'mp4',
 406                 'title': 'Охотское море стало целиком российским',
 407                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 408             },
 409             'params': {
 410                 # m3u8 download
 411                 'skip_download': True,
 412             },
 413         },
 414         # TVC embed
 415         {
 416             '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/',
 417             'info_dict': {
 418                 'id': '55304',
 419                 'ext': 'mp4',
 420                 'title': 'Дошкольное воспитание',
 421             },
 422         },
 423         # SportBox embed
 424         {
 425             'url': 'http://www.vestifinance.ru/articles/25753',
 426             'info_dict': {
 427                 'id': '25753',
 428                 'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
 429             },
 430             'playlist': [{
 431                 'info_dict': {
 432                     'id': '370908',
 433                     'title': 'Госзаказ. День 3',
 434                     'ext': 'mp4',
 435                 }
 436             }, {
 437                 'info_dict': {
 438                     'id': '370905',
 439                     'title': 'Госзаказ. День 2',
 440                     'ext': 'mp4',
 441                 }
 442             }, {
 443                 'info_dict': {
 444                     'id': '370902',
 445                     'title': 'Госзаказ. День 1',
 446                     'ext': 'mp4',
 447                 }
 448             }],
 449             'params': {
 450                 # m3u8 download
 451                 'skip_download': True,
 452             },
 453         },
 454         # Myvi.ru embed
 455         {
 456             'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
 457             'info_dict': {
 458                 'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
 459                 'ext': 'mp4',
 460                 'title': 'Ужастики, русский трейлер (2015)',
 461                 'thumbnail': 're:^https?://.*\.jpg$',
 462                 'duration': 153,
 463             }
 464         },
 465         # XHamster embed
 466         {
 467             'url': 'http://www.numisc.com/forum/showthread.php?11696-FM15-which-pumiscer-was-this-%28-vid-%29-%28-alfa-as-fuck-srx-%29&s=711f5db534502e22260dec8c5e2d66d8',
 468             'info_dict': {
 469                 'id': 'showthread',
 470                 'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
 471             },
 472             'playlist_mincount': 7,
 473         },
 474         # Embedded TED video
 475         {
 476             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 477             'md5': '65fdff94098e4a607385a60c5177c638',
 478             'info_dict': {
 479                 'id': '1969',
 480                 'ext': 'mp4',
 481                 'title': 'Hidden miracles of the natural world',
 482                 'uploader': 'Louie Schwartzberg',
 483                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 484             }
 485         },
 486         # Embeded Ustream video
 487         {
 488             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 489             'md5': '27b99cdb639c9b12a79bca876a073417',
 490             'info_dict': {
 491                 'id': '45734260',
 492                 'ext': 'flv',
 493                 'uploader': 'AU SPA:  The NSA and Privacy',
 494                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 495             }
 496         },
 497         # nowvideo embed hidden behind percent encoding
 498         {
 499             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 500             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 501             'info_dict': {
 502                 'id': '06e53103ca9aa',
 503                 'ext': 'flv',
 504                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 505                 'description': 'No description',
 506             },
 507         },
 508         # arte embed
 509         {
 510             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 511             'md5': '7653032cbb25bf6c80d80f217055fa43',
 512             'info_dict': {
 513                 'id': '048195-004_PLUS7-F',
 514                 'ext': 'flv',
 515                 'title': 'X:enius',
 516                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 517                 'upload_date': '20140320',
 518             },
 519             'params': {
 520                 'skip_download': 'Requires rtmpdump'
 521             }
 522         },
 523         # francetv embed
 524         {
 525             'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
 526             'info_dict': {
 527                 'id': 'EV_30231',
 528                 'ext': 'mp4',
 529                 'title': 'Alcaline, le concert avec Calogero',
 530                 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
 531                 'upload_date': '20150226',
 532                 'timestamp': 1424989860,
 533                 'duration': 5400,
 534             },
 535             'params': {
 536                 # m3u8 downloads
 537                 'skip_download': True,
 538             },
 539             'expected_warnings': [
 540                 'Forbidden'
 541             ]
 542         },
 543         # Condé Nast embed
 544         {
 545             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 546             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 547             'info_dict': {
 548                 'id': '53501be369702d3275860000',
 549                 'ext': 'mp4',
 550                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 551             }
 552         },
 553         # Dailymotion embed
 554         {
 555             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 556             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 557             'info_dict': {
 558                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 559                 'ext': 'mp4',
 560                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 561                 'uploader': 'Spi0n',
 562             },
 563             'add_ie': ['Dailymotion'],
 564         },
 565         # YouTube embed
 566         {
 567             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 568             'info_dict': {
 569                 'id': 'FXRb4ykk4S0',
 570                 'ext': 'mp4',
 571                 'title': 'The NBL Auction 2014',
 572                 'uploader': 'BADMINTON England',
 573                 'uploader_id': 'BADMINTONEvents',
 574                 'upload_date': '20140603',
 575                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 576             },
 577             'add_ie': ['Youtube'],
 578             'params': {
 579                 'skip_download': True,
 580             }
 581         },
 582         # MTVSercices embed
 583         {
 584             'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
 585             'md5': '35727f82f58c76d996fc188f9755b0d5',
 586             'info_dict': {
 587                 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
 588                 'ext': 'mp4',
 589                 'title': 'Review',
 590                 'description': 'Mario\'s life in the fast lane has never looked so good.',
 591             },
 592         },
 593         # YouTube embed via <data-embed-url="">
 594         {
 595             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 596             'info_dict': {
 597                 'id': '4vAffPZIT44',
 598                 'ext': 'mp4',
 599                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 600                 'uploader': 'Gameloft',
 601                 'uploader_id': 'gameloft',
 602                 'upload_date': '20140828',
 603                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 604             },
 605             'params': {
 606                 'skip_download': True,
 607             }
 608         },
 609         # Camtasia studio
 610         {
 611             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 612             'playlist': [{
 613                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 614                 'info_dict': {
 615                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 616                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 617                     'ext': 'flv',
 618                     'duration': 2235.90,
 619                 }
 620             }, {
 621                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 622                 'info_dict': {
 623                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 624                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 625                     'ext': 'flv',
 626                     'duration': 2235.93,
 627                 }
 628             }],
 629             'info_dict': {
 630                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 631             }
 632         },
 633         # Flowplayer
 634         {
 635             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 636             'md5': '9d65602bf31c6e20014319c7d07fba27',
 637             'info_dict': {
 638                 'id': '5123ea6d5e5a7',
 639                 'ext': 'mp4',
 640                 'age_limit': 18,
 641                 'uploader': 'www.handjobhub.com',
 642                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 643             }
 644         },
 645         # Multiple brightcove videos
 646         # https://github.com/rg3/youtube-dl/issues/2283
 647         {
 648             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 649             'info_dict': {
 650                 'id': 'always-never',
 651                 'title': 'Always / Never - The New Yorker',
 652             },
 653             'playlist_count': 3,
 654             'params': {
 655                 'extract_flat': False,
 656                 'skip_download': True,
 657             }
 658         },
 659         # MLB embed
 660         {
 661             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 662             'md5': '96f09a37e44da40dd083e12d9a683327',
 663             'info_dict': {
 664                 'id': '33322633',
 665                 'ext': 'mp4',
 666                 'title': 'Ump changes call to ball',
 667                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 668                 'duration': 48,
 669                 'timestamp': 1401537900,
 670                 'upload_date': '20140531',
 671                 'thumbnail': 're:^https?://.*\.jpg$',
 672             },
 673         },
 674         # Wistia embed
 675         {
 676             'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 677             'md5': '8788b683c777a5cf25621eaf286d0c23',
 678             'info_dict': {
 679                 'id': '1cfaf6b7ea',
 680                 'ext': 'mov',
 681                 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
 682                 'duration': 643.0,
 683                 'filesize': 182808282,
 684                 'uploader': 'education-portal.com',
 685             },
 686         },
 687         {
 688             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 689             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 690             'info_dict': {
 691                 'id': 'uxjb0lwrcz',
 692                 'ext': 'mp4',
 693                 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
 694                 'duration': 1715.0,
 695                 'uploader': 'thoughtworks.wistia.com',
 696             },
 697         },
 698         # Soundcloud embed
 699         {
 700             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 701             'info_dict': {
 702                 'id': '174391317',
 703                 'ext': 'mp3',
 704                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 705                 'uploader': 'Sophos Security',
 706                 'title': 'Chet Chat 171 - Oct 29, 2014',
 707                 'upload_date': '20141029',
 708             }
 709         },
 710         # Livestream embed
 711         {
 712             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 713             'info_dict': {
 714                 'id': '67864563',
 715                 'ext': 'flv',
 716                 'upload_date': '20141112',
 717                 'title': 'Rosetta #CometLanding webcast HL 10',
 718             }
 719         },
 720         # LazyYT
 721         {
 722             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
 723             'info_dict': {
 724                 'id': '1986',
 725                 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
 726             },
 727             'playlist_mincount': 2,
 728         },
 729         # Cinchcast embed
 730         {
 731             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
 732             'info_dict': {
 733                 'id': '7141703',
 734                 'ext': 'mp3',
 735                 'upload_date': '20141126',
 736                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
 737             }
 738         },
 739         # Cinerama player
 740         {
 741             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
 742             'info_dict': {
 743                 'id': '730m_DandD_1901_512k',
 744                 'ext': 'mp4',
 745                 'uploader': 'www.abc.net.au',
 746                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
 747             }
 748         },
 749         # embedded viddler video
 750         {
 751             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
 752             'info_dict': {
 753                 'id': '4d03aad9',
 754                 'ext': 'mp4',
 755                 'uploader': 'deadspin',
 756                 'title': 'WALL-TO-GORTAT',
 757                 'timestamp': 1422285291,
 758                 'upload_date': '20150126',
 759             },
 760             'add_ie': ['Viddler'],
 761         },
 762         # Libsyn embed
 763         {
 764             'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
 765             'info_dict': {
 766                 'id': '3377616',
 767                 'ext': 'mp3',
 768                 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
 769                 'description': 'md5:601cb790edd05908957dae8aaa866465',
 770                 'upload_date': '20150220',
 771             },
 772         },
 773         # jwplayer YouTube
 774         {
 775             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
 776             'info_dict': {
 777                 'id': 'Mrj4DVp2zeA',
 778                 'ext': 'mp4',
 779                 'upload_date': '20150212',
 780                 'uploader': 'The National Archives UK',
 781                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
 782                 'uploader_id': 'NationalArchives08',
 783                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
 784             },
 785         },
 786         # rtl.nl embed
 787         {
 788             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
 789             'playlist_mincount': 5,
 790             'info_dict': {
 791                 'id': 'aanslagen-kopenhagen',
 792                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
 793             }
 794         },
 795         # Zapiks embed
 796         {
 797             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
 798             'info_dict': {
 799                 'id': '118046',
 800                 'ext': 'mp4',
 801                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
 802             }
 803         },
 804         # Kaltura embed
 805         {
 806             'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
 807             'info_dict': {
 808                 'id': '1_eergr3h1',
 809                 'ext': 'mp4',
 810                 'upload_date': '20150226',
 811                 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
 812                 'timestamp': int,
 813                 'title': 'John Carlson Postgame 2/25/15',
 814             },
 815         },
 816         # Kaltura embed (different embed code)
 817         {
 818             'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
 819             'info_dict': {
 820                 'id': '1_a52wc67y',
 821                 'ext': 'flv',
 822                 'upload_date': '20150127',
 823                 'uploader_id': 'PremierMedia',
 824                 'timestamp': int,
 825                 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
 826             },
 827         },
 828         # Kaltura embed protected with referrer
 829         {
 830             'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero',
 831             'info_dict': {
 832                 'id': '1_g4fbemnq',
 833                 'ext': 'mp4',
 834                 'title': 'Violetta - Achter De Schermen - Ruggero',
 835                 'description': 'Achter de schermen met Ruggero',
 836                 'timestamp': 1435133761,
 837                 'upload_date': '20150624',
 838                 'uploader_id': 'echojecka',
 839             },
 840         },
 841         # Eagle.Platform embed (generic URL)
 842         {
 843             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
 844             'info_dict': {
 845                 'id': '227304',
 846                 'ext': 'mp4',
 847                 'title': 'Навальный вышел на свободу',
 848                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
 849                 'thumbnail': 're:^https?://.*\.jpg$',
 850                 'duration': 87,
 851                 'view_count': int,
 852                 'age_limit': 0,
 853             },
 854         },
 855         # ClipYou (Eagle.Platform) embed (custom URL)
 856         {
 857             'url': 'http://muz-tv.ru/play/7129/',
 858             'info_dict': {
 859                 'id': '12820',
 860                 'ext': 'mp4',
 861                 'title': "'O Sole Mio",
 862                 'thumbnail': 're:^https?://.*\.jpg$',
 863                 'duration': 216,
 864                 'view_count': int,
 865             },
 866         },
 867         # Pladform embed
 868         {
 869             'url': 'http://muz-tv.ru/kinozal/view/7400/',
 870             'info_dict': {
 871                 'id': '100183293',
 872                 'ext': 'mp4',
 873                 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
 874                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
 875                 'thumbnail': 're:^https?://.*\.jpg$',
 876                 'duration': 694,
 877                 'age_limit': 0,
 878             },
 879         },
 880         # Playwire embed
 881         {
 882             'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
 883             'info_dict': {
 884                 'id': '3519514',
 885                 'ext': 'mp4',
 886                 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
 887                 'thumbnail': 're:^https?://.*\.png$',
 888                 'duration': 45.115,
 889             },
 890         },
 891         # 5min embed
 892         {
 893             'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
 894             'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
 895             'info_dict': {
 896                 'id': '518726732',
 897                 'ext': 'mp4',
 898                 'title': 'Facebook Creates "On This Day" | Crunch Report',
 899             },
 900         },
 901         # SVT embed
 902         {
 903             'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 904             'info_dict': {
 905                 'id': '2900353',
 906                 'ext': 'flv',
 907                 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
 908                 'duration': 27,
 909                 'age_limit': 0,
 910             },
 911         },
 912         # Crooks and Liars embed
 913         {
 914             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
 915             'info_dict': {
 916                 'id': '8RUoRhRi',
 917                 'ext': 'mp4',
 918                 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
 919                 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
 920                 'timestamp': 1428207000,
 921                 'upload_date': '20150405',
 922                 'uploader': 'Heather',
 923             },
 924         },
 925         # Crooks and Liars external embed
 926         {
 927             'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
 928             'info_dict': {
 929                 'id': 'MTE3MjUtMzQ2MzA',
 930                 'ext': 'mp4',
 931                 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
 932                 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
 933                 'timestamp': 1265032391,
 934                 'upload_date': '20100201',
 935                 'uploader': 'Heather',
 936             },
 937         },
 938         # NBC Sports vplayer embed
 939         {
 940             'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
 941             'info_dict': {
 942                 'id': 'ln7x1qSThw4k',
 943                 'ext': 'flv',
 944                 'title': "PFT Live: New leader in the 'new-look' defense",
 945                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
 946             },
 947         },
 948         # UDN embed
 949         {
 950             'url': 'http://www.udn.com/news/story/7314/822787',
 951             'md5': 'fd2060e988c326991037b9aff9df21a6',
 952             'info_dict': {
 953                 'id': '300346',
 954                 'ext': 'mp4',
 955                 'title': '中一中男師變性 全校師生力挺',
 956                 'thumbnail': 're:^https?://.*\.jpg$',
 957             }
 958         },
 959         # Ooyala embed
 960         {
 961             'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
 962             'info_dict': {
 963                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
 964                 'ext': 'mp4',
 965                 'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.',
 966                 'title': 'This is what separates the Excel masters from the wannabes',
 967                 'duration': 191933,
 968             },
 969             'params': {
 970                 # m3u8 downloads
 971                 'skip_download': True,
 972             }
 973         },
 974         # Contains a SMIL manifest
 975         {
 976             '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',
 977             'info_dict': {
 978                 'id': 'file',
 979                 'ext': 'flv',
 980                 'title': '+ Football: Lottery Champions League Europe',
 981                 'uploader': 'www.telewebion.com',
 982             },
 983             'params': {
 984                 # rtmpe downloads
 985                 'skip_download': True,
 986             }
 987         },
 988         # Brightcove URL in single quotes
 989         {
 990             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
 991             'md5': '4ae374f1f8b91c889c4b9203c8c752af',
 992             'info_dict': {
 993                 'id': '4255764656001',
 994                 'ext': 'mp4',
 995                 'title': 'SN Presents: Russell Martin, World Citizen',
 996                 '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.',
 997                 'uploader': 'Rogers Sportsnet',
 998             },
 999         },
1000         # Dailymotion Cloud video
1001         {
1002             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
1003             'md5': '49444254273501a64675a7e68c502681',
1004             'info_dict': {
1005                 'id': '5585de919473990de4bee11b',
1006                 'ext': 'mp4',
1007                 'title': 'Le débat',
1008                 'thumbnail': 're:^https?://.*\.jpe?g$',
1009             }
1010         },
1011         # OnionStudios embed
1012         {
1013             'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
1014             'info_dict': {
1015                 'id': '2855',
1016                 'ext': 'mp4',
1017                 'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
1018                 'thumbnail': 're:^https?://.*\.jpe?g$',
1019                 'uploader': 'ClickHole',
1020                 'uploader_id': 'clickhole',
1021             }
1022         },
1023         # SnagFilms embed
1024         {
1025             'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
1026             'info_dict': {
1027                 'id': '74849a00-85a9-11e1-9660-123139220831',
1028                 'ext': 'mp4',
1029                 'title': '#whilewewatch',
1030             }
1031         },
1032         # AdobeTVVideo embed
1033         {
1034             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
1035             'md5': '43662b577c018ad707a63766462b1e87',
1036             'info_dict': {
1037                 'id': '2456',
1038                 'ext': 'mp4',
1039                 'title': 'New experience with Acrobat DC',
1040                 'description': 'New experience with Acrobat DC',
1041                 'duration': 248.667,
1042             },
1043         },
1044         # ScreenwaveMedia embed
1045         {
1046             'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1',
1047             'md5': '24ace5baba0d35d55c6810b51f34e9e0',
1048             'info_dict': {
1049                 'id': 'cinemasnob-55d26273809dd',
1050                 'ext': 'mp4',
1051                 'title': 'cinemasnob',
1052             },
1053         },
1054         # BrightcoveInPageEmbed embed
1055         {
1056             'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
1057             'info_dict': {
1058                 'id': '4238694884001',
1059                 'ext': 'flv',
1060                 'title': 'Tabletop: Dread, Last Thoughts',
1061                 'description': 'Tabletop: Dread, Last Thoughts',
1062                 'duration': 51690,
1063             },
1064         },
1065         # JWPlayer with M3U8
1066         {
1067             'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video',
1068             'info_dict': {
1069                 'id': 'playlist',
1070                 'ext': 'mp4',
1071                 'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ',
1072                 'uploader': 'ren.tv',
1073             },
1074             'params': {
1075                 # m3u8 downloads
1076                 'skip_download': True,
1077             }
1078         }
1079     ]
1080
1081     def report_following_redirect(self, new_url):
1082         """Report information extraction."""
1083         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
1084
1085     def _extract_rss(self, url, video_id, doc):
1086         playlist_title = doc.find('./channel/title').text
1087         playlist_desc_el = doc.find('./channel/description')
1088         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
1089
1090         entries = []
1091         for it in doc.findall('./channel/item'):
1092             next_url = xpath_text(it, 'link', fatal=False)
1093             if not next_url:
1094                 enclosure_nodes = it.findall('./enclosure')
1095                 for e in enclosure_nodes:
1096                     next_url = e.attrib.get('url')
1097                     if next_url:
1098                         break
1099
1100             if not next_url:
1101                 continue
1102
1103             entries.append({
1104                 '_type': 'url',
1105                 'url': next_url,
1106                 'title': it.find('title').text,
1107             })
1108
1109         return {
1110             '_type': 'playlist',
1111             'id': url,
1112             'title': playlist_title,
1113             'description': playlist_desc,
1114             'entries': entries,
1115         }
1116
1117     def _extract_camtasia(self, url, video_id, webpage):
1118         """ Returns None if no camtasia video can be found. """
1119
1120         camtasia_cfg = self._search_regex(
1121             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
1122             webpage, 'camtasia configuration file', default=None)
1123         if camtasia_cfg is None:
1124             return None
1125
1126         title = self._html_search_meta('DC.title', webpage, fatal=True)
1127
1128         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
1129         camtasia_cfg = self._download_xml(
1130             camtasia_url, video_id,
1131             note='Downloading camtasia configuration',
1132             errnote='Failed to download camtasia configuration')
1133         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
1134
1135         entries = []
1136         for n in fileset_node.getchildren():
1137             url_n = n.find('./uri')
1138             if url_n is None:
1139                 continue
1140
1141             entries.append({
1142                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
1143                 'title': '%s - %s' % (title, n.tag),
1144                 'url': compat_urlparse.urljoin(url, url_n.text),
1145                 'duration': float_or_none(n.find('./duration').text),
1146             })
1147
1148         return {
1149             '_type': 'playlist',
1150             'entries': entries,
1151             'title': title,
1152         }
1153
1154     def _real_extract(self, url):
1155         if url.startswith('//'):
1156             return {
1157                 '_type': 'url',
1158                 'url': self.http_scheme() + url,
1159             }
1160
1161         parsed_url = compat_urlparse.urlparse(url)
1162         if not parsed_url.scheme:
1163             default_search = self._downloader.params.get('default_search')
1164             if default_search is None:
1165                 default_search = 'fixup_error'
1166
1167             if default_search in ('auto', 'auto_warning', 'fixup_error'):
1168                 if '/' in url:
1169                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
1170                     return self.url_result('http://' + url)
1171                 elif default_search != 'fixup_error':
1172                     if default_search == 'auto_warning':
1173                         if re.match(r'^(?:url|URL)$', url):
1174                             raise ExtractorError(
1175                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
1176                                 expected=True)
1177                         else:
1178                             self._downloader.report_warning(
1179                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
1180                     return self.url_result('ytsearch:' + url)
1181
1182             if default_search in ('error', 'fixup_error'):
1183                 raise ExtractorError(
1184                     '%r is not a valid URL. '
1185                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
1186                     % (url, url), expected=True)
1187             else:
1188                 if ':' not in default_search:
1189                     default_search += ':'
1190                 return self.url_result(default_search + url)
1191
1192         url, smuggled_data = unsmuggle_url(url)
1193         force_videoid = None
1194         is_intentional = smuggled_data and smuggled_data.get('to_generic')
1195         if smuggled_data and 'force_videoid' in smuggled_data:
1196             force_videoid = smuggled_data['force_videoid']
1197             video_id = force_videoid
1198         else:
1199             video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
1200
1201         self.to_screen('%s: Requesting header' % video_id)
1202
1203         head_req = HEADRequest(url)
1204         head_response = self._request_webpage(
1205             head_req, video_id,
1206             note=False, errnote='Could not send HEAD request to %s' % url,
1207             fatal=False)
1208
1209         if head_response is not False:
1210             # Check for redirect
1211             new_url = head_response.geturl()
1212             if url != new_url:
1213                 self.report_following_redirect(new_url)
1214                 if force_videoid:
1215                     new_url = smuggle_url(
1216                         new_url, {'force_videoid': force_videoid})
1217                 return self.url_result(new_url)
1218
1219         full_response = None
1220         if head_response is False:
1221             request = sanitized_Request(url)
1222             request.add_header('Accept-Encoding', '*')
1223             full_response = self._request_webpage(request, video_id)
1224             head_response = full_response
1225
1226         # Check for direct link to a video
1227         content_type = head_response.headers.get('Content-Type', '')
1228         m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
1229         if m:
1230             upload_date = unified_strdate(
1231                 head_response.headers.get('Last-Modified'))
1232             return {
1233                 'id': video_id,
1234                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1235                 'direct': True,
1236                 'formats': [{
1237                     'format_id': m.group('format_id'),
1238                     'url': url,
1239                     'vcodec': 'none' if m.group('type') == 'audio' else None
1240                 }],
1241                 'upload_date': upload_date,
1242             }
1243
1244         if not self._downloader.params.get('test', False) and not is_intentional:
1245             force = self._downloader.params.get('force_generic_extractor', False)
1246             self._downloader.report_warning(
1247                 '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
1248
1249         if not full_response:
1250             request = sanitized_Request(url)
1251             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
1252             # making it impossible to download only chunk of the file (yet we need only 512kB to
1253             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
1254             # that will always result in downloading the whole file that is not desirable.
1255             # Therefore for extraction pass we have to override Accept-Encoding to any in order
1256             # to accept raw bytes and being able to download only a chunk.
1257             # It may probably better to solve this by checking Content-Type for application/octet-stream
1258             # after HEAD request finishes, but not sure if we can rely on this.
1259             request.add_header('Accept-Encoding', '*')
1260             full_response = self._request_webpage(request, video_id)
1261
1262         # Maybe it's a direct link to a video?
1263         # Be careful not to download the whole thing!
1264         first_bytes = full_response.read(512)
1265         if not is_html(first_bytes):
1266             self._downloader.report_warning(
1267                 'URL could be a direct video link, returning it as such.')
1268             upload_date = unified_strdate(
1269                 head_response.headers.get('Last-Modified'))
1270             return {
1271                 'id': video_id,
1272                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1273                 'direct': True,
1274                 'url': url,
1275                 'upload_date': upload_date,
1276             }
1277
1278         webpage = self._webpage_read_content(
1279             full_response, url, video_id, prefix=first_bytes)
1280
1281         self.report_extraction(video_id)
1282
1283         # Is it an RSS feed, a SMIL file or a XSPF playlist?
1284         try:
1285             doc = compat_etree_fromstring(webpage.encode('utf-8'))
1286             if doc.tag == 'rss':
1287                 return self._extract_rss(url, video_id, doc)
1288             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
1289                 return self._parse_smil(doc, url, video_id)
1290             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
1291                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
1292         except compat_xml_parse_error:
1293             pass
1294
1295         # Is it a Camtasia project?
1296         camtasia_res = self._extract_camtasia(url, video_id, webpage)
1297         if camtasia_res is not None:
1298             return camtasia_res
1299
1300         # Sometimes embedded video player is hidden behind percent encoding
1301         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1302         # Unescaping the whole page allows to handle those cases in a generic way
1303         webpage = compat_urllib_parse_unquote(webpage)
1304
1305         # it's tempting to parse this further, but you would
1306         # have to take into account all the variations like
1307         #   Video Title - Site Name
1308         #   Site Name | Video Title
1309         #   Video Title - Tagline | Site Name
1310         # and so on and so forth; it's just not practical
1311         video_title = self._html_search_regex(
1312             r'(?s)<title>(.*?)</title>', webpage, 'video title',
1313             default='video')
1314
1315         # Try to detect age limit automatically
1316         age_limit = self._rta_search(webpage)
1317         # And then there are the jokers who advertise that they use RTA,
1318         # but actually don't.
1319         AGE_LIMIT_MARKERS = [
1320             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1321         ]
1322         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1323             age_limit = 18
1324
1325         # video uploader is domain name
1326         video_uploader = self._search_regex(
1327             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1328
1329         # Helper method
1330         def _playlist_from_matches(matches, getter=None, ie=None):
1331             urlrs = orderedSet(
1332                 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1333                 for m in matches)
1334             return self.playlist_result(
1335                 urlrs, playlist_id=video_id, playlist_title=video_title)
1336
1337         # Look for Brightcove Legacy Studio embeds
1338         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
1339         if bc_urls:
1340             self.to_screen('Brightcove video detected.')
1341             entries = [{
1342                 '_type': 'url',
1343                 'url': smuggle_url(bc_url, {'Referer': url}),
1344                 'ie_key': 'BrightcoveLegacy'
1345             } for bc_url in bc_urls]
1346
1347             return {
1348                 '_type': 'playlist',
1349                 'title': video_title,
1350                 'id': video_id,
1351                 'entries': entries,
1352             }
1353
1354         # Look for Brightcove New Studio embeds
1355         bc_urls = BrightcoveNewIE._extract_urls(webpage)
1356         if bc_urls:
1357             return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
1358
1359         # Look for embedded rtl.nl player
1360         matches = re.findall(
1361             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1362             webpage)
1363         if matches:
1364             return _playlist_from_matches(matches, ie='RtlNl')
1365
1366         vimeo_url = VimeoIE._extract_vimeo_url(url, webpage)
1367         if vimeo_url is not None:
1368             return self.url_result(vimeo_url)
1369
1370         vid_me_embed_url = self._search_regex(
1371             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
1372             webpage, 'vid.me embed', default=None)
1373         if vid_me_embed_url is not None:
1374             return self.url_result(vid_me_embed_url, 'Vidme')
1375
1376         # Look for embedded YouTube player
1377         matches = re.findall(r'''(?x)
1378             (?:
1379                 <iframe[^>]+?src=|
1380                 data-video-url=|
1381                 <embed[^>]+?src=|
1382                 embedSWF\(?:\s*|
1383                 new\s+SWFObject\(
1384             )
1385             (["\'])
1386                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1387                 (?:embed|v|p)/.+?)
1388             \1''', webpage)
1389         if matches:
1390             return _playlist_from_matches(
1391                 matches, lambda m: unescapeHTML(m[1]))
1392
1393         # Look for lazyYT YouTube embed
1394         matches = re.findall(
1395             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1396         if matches:
1397             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1398
1399         # Look for embedded Dailymotion player
1400         matches = re.findall(
1401             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
1402         if matches:
1403             return _playlist_from_matches(
1404                 matches, lambda m: unescapeHTML(m[1]))
1405
1406         # Look for embedded Dailymotion playlist player (#3822)
1407         m = re.search(
1408             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1409         if m:
1410             playlists = re.findall(
1411                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1412             if playlists:
1413                 return _playlist_from_matches(
1414                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1415
1416         # Look for embedded Wistia player
1417         match = re.search(
1418             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1419         if match:
1420             embed_url = self._proto_relative_url(
1421                 unescapeHTML(match.group('url')))
1422             return {
1423                 '_type': 'url_transparent',
1424                 'url': embed_url,
1425                 'ie_key': 'Wistia',
1426                 'uploader': video_uploader,
1427                 'title': video_title,
1428                 'id': video_id,
1429             }
1430
1431         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1432         if match:
1433             return {
1434                 '_type': 'url_transparent',
1435                 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1436                 'ie_key': 'Wistia',
1437                 'uploader': video_uploader,
1438                 'title': video_title,
1439                 'id': match.group('id')
1440             }
1441
1442         # Look for embedded blip.tv player
1443         bliptv_url = BlipTVIE._extract_url(webpage)
1444         if bliptv_url:
1445             return self.url_result(bliptv_url, 'BlipTV')
1446
1447         # Look for SVT player
1448         svt_url = SVTIE._extract_url(webpage)
1449         if svt_url:
1450             return self.url_result(svt_url, 'SVT')
1451
1452         # Look for embedded condenast player
1453         matches = re.findall(
1454             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1455             webpage)
1456         if matches:
1457             return {
1458                 '_type': 'playlist',
1459                 'entries': [{
1460                     '_type': 'url',
1461                     'ie_key': 'CondeNast',
1462                     'url': ma,
1463                 } for ma in matches],
1464                 'title': video_title,
1465                 'id': video_id,
1466             }
1467
1468         # Look for Bandcamp pages with custom domain
1469         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1470         if mobj is not None:
1471             burl = unescapeHTML(mobj.group(1))
1472             # Don't set the extractor because it can be a track url or an album
1473             return self.url_result(burl)
1474
1475         # Look for embedded Vevo player
1476         mobj = re.search(
1477             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1478         if mobj is not None:
1479             return self.url_result(mobj.group('url'))
1480
1481         # Look for embedded Viddler player
1482         mobj = re.search(
1483             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1484             webpage)
1485         if mobj is not None:
1486             return self.url_result(mobj.group('url'))
1487
1488         # Look for NYTimes player
1489         mobj = re.search(
1490             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1491             webpage)
1492         if mobj is not None:
1493             return self.url_result(mobj.group('url'))
1494
1495         # Look for Libsyn player
1496         mobj = re.search(
1497             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1498         if mobj is not None:
1499             return self.url_result(mobj.group('url'))
1500
1501         # Look for Ooyala videos
1502         mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1503                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1504                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1505                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1506         if mobj is not None:
1507             return OoyalaIE._build_url_result(smuggle_url(mobj.group('ec'), {'domain': url}))
1508
1509         # Look for multiple Ooyala embeds on SBN network websites
1510         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1511         if mobj is not None:
1512             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1513             if embeds:
1514                 return _playlist_from_matches(
1515                     embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
1516
1517         # Look for Aparat videos
1518         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1519         if mobj is not None:
1520             return self.url_result(mobj.group(1), 'Aparat')
1521
1522         # Look for MPORA videos
1523         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1524         if mobj is not None:
1525             return self.url_result(mobj.group(1), 'Mpora')
1526
1527         # Look for embedded NovaMov-based player
1528         mobj = re.search(
1529             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1530                     (?P<url>http://(?:(?:embed|www)\.)?
1531                         (?:novamov\.com|
1532                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
1533                            videoweed\.(?:es|com)|
1534                            movshare\.(?:net|sx|ag)|
1535                            divxstage\.(?:eu|net|ch|co|at|ag))
1536                         /embed\.php.+?)\1''', webpage)
1537         if mobj is not None:
1538             return self.url_result(mobj.group('url'))
1539
1540         # Look for embedded Facebook player
1541         mobj = re.search(
1542             r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1543         if mobj is not None:
1544             return self.url_result(mobj.group('url'), 'Facebook')
1545
1546         # Look for embedded VK player
1547         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1548         if mobj is not None:
1549             return self.url_result(mobj.group('url'), 'VK')
1550
1551         # Look for embedded ivi player
1552         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1553         if mobj is not None:
1554             return self.url_result(mobj.group('url'), 'Ivi')
1555
1556         # Look for embedded Huffington Post player
1557         mobj = re.search(
1558             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1559         if mobj is not None:
1560             return self.url_result(mobj.group('url'), 'HuffPost')
1561
1562         # Look for embed.ly
1563         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1564         if mobj is not None:
1565             return self.url_result(mobj.group('url'))
1566         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1567         if mobj is not None:
1568             return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
1569
1570         # Look for funnyordie embed
1571         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1572         if matches:
1573             return _playlist_from_matches(
1574                 matches, getter=unescapeHTML, ie='FunnyOrDie')
1575
1576         # Look for BBC iPlayer embed
1577         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1578         if matches:
1579             return _playlist_from_matches(matches, ie='BBCCoUk')
1580
1581         # Look for embedded RUTV player
1582         rutv_url = RUTVIE._extract_url(webpage)
1583         if rutv_url:
1584             return self.url_result(rutv_url, 'RUTV')
1585
1586         # Look for embedded TVC player
1587         tvc_url = TVCIE._extract_url(webpage)
1588         if tvc_url:
1589             return self.url_result(tvc_url, 'TVC')
1590
1591         # Look for embedded SportBox player
1592         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1593         if sportbox_urls:
1594             return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1595
1596         # Look for embedded PornHub player
1597         pornhub_url = PornHubIE._extract_url(webpage)
1598         if pornhub_url:
1599             return self.url_result(pornhub_url, 'PornHub')
1600
1601         # Look for embedded XHamster player
1602         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
1603         if xhamster_urls:
1604             return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
1605
1606         # Look for embedded Tvigle player
1607         mobj = re.search(
1608             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
1609         if mobj is not None:
1610             return self.url_result(mobj.group('url'), 'Tvigle')
1611
1612         # Look for embedded TED player
1613         mobj = re.search(
1614             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1615         if mobj is not None:
1616             return self.url_result(mobj.group('url'), 'TED')
1617
1618         # Look for embedded Ustream videos
1619         mobj = re.search(
1620             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1621         if mobj is not None:
1622             return self.url_result(mobj.group('url'), 'Ustream')
1623
1624         # Look for embedded arte.tv player
1625         mobj = re.search(
1626             r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1627             webpage)
1628         if mobj is not None:
1629             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1630
1631         # Look for embedded francetv player
1632         mobj = re.search(
1633             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
1634             webpage)
1635         if mobj is not None:
1636             return self.url_result(mobj.group('url'))
1637
1638         # Look for embedded smotri.com player
1639         smotri_url = SmotriIE._extract_url(webpage)
1640         if smotri_url:
1641             return self.url_result(smotri_url, 'Smotri')
1642
1643         # Look for embedded Myvi.ru player
1644         myvi_url = MyviIE._extract_url(webpage)
1645         if myvi_url:
1646             return self.url_result(myvi_url)
1647
1648         # Look for embeded soundcloud player
1649         mobj = re.search(
1650             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1651             webpage)
1652         if mobj is not None:
1653             url = unescapeHTML(mobj.group('url'))
1654             return self.url_result(url)
1655
1656         # Look for embedded vulture.com player
1657         mobj = re.search(
1658             r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1659             webpage)
1660         if mobj is not None:
1661             url = unescapeHTML(mobj.group('url'))
1662             return self.url_result(url, ie='Vulture')
1663
1664         # Look for embedded mtvservices player
1665         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
1666         if mtvservices_url:
1667             return self.url_result(mtvservices_url, ie='MTVServicesEmbedded')
1668
1669         # Look for embedded yahoo player
1670         mobj = re.search(
1671             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1672             webpage)
1673         if mobj is not None:
1674             return self.url_result(mobj.group('url'), 'Yahoo')
1675
1676         # Look for embedded sbs.com.au player
1677         mobj = re.search(
1678             r'''(?x)
1679             (?:
1680                 <meta\s+property="og:video"\s+content=|
1681                 <iframe[^>]+?src=
1682             )
1683             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1684             webpage)
1685         if mobj is not None:
1686             return self.url_result(mobj.group('url'), 'SBS')
1687
1688         # Look for embedded Cinchcast player
1689         mobj = re.search(
1690             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1691             webpage)
1692         if mobj is not None:
1693             return self.url_result(mobj.group('url'), 'Cinchcast')
1694
1695         mobj = re.search(
1696             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1697             webpage)
1698         if not mobj:
1699             mobj = re.search(
1700                 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
1701                 webpage)
1702         if mobj is not None:
1703             return self.url_result(mobj.group('url'), 'MLB')
1704
1705         mobj = re.search(
1706             r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1707             webpage)
1708         if mobj is not None:
1709             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1710
1711         mobj = re.search(
1712             r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1713             webpage)
1714         if mobj is not None:
1715             return self.url_result(mobj.group('url'), 'Livestream')
1716
1717         # Look for Zapiks embed
1718         mobj = re.search(
1719             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1720         if mobj is not None:
1721             return self.url_result(mobj.group('url'), 'Zapiks')
1722
1723         # Look for Kaltura embeds
1724         mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or
1725                 re.search(r'(?s)(?P<q1>["\'])(?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?(?P=q1).*?entry_?[Ii]d\s*:\s*(?P<q2>["\'])(?P<id>.+?)(?P=q2)', webpage))
1726         if mobj is not None:
1727             return self.url_result(smuggle_url(
1728                 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(),
1729                 {'source_url': url}), 'Kaltura')
1730
1731         # Look for Eagle.Platform embeds
1732         mobj = re.search(
1733             r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1734         if mobj is not None:
1735             return self.url_result(mobj.group('url'), 'EaglePlatform')
1736
1737         # Look for ClipYou (uses Eagle.Platform) embeds
1738         mobj = re.search(
1739             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1740         if mobj is not None:
1741             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1742
1743         # Look for Pladform embeds
1744         mobj = re.search(
1745             r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
1746         if mobj is not None:
1747             return self.url_result(mobj.group('url'), 'Pladform')
1748
1749         # Look for Playwire embeds
1750         mobj = re.search(
1751             r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
1752         if mobj is not None:
1753             return self.url_result(mobj.group('url'))
1754
1755         # Look for 5min embeds
1756         mobj = re.search(
1757             r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1758         if mobj is not None:
1759             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1760
1761         # Look for Crooks and Liars embeds
1762         mobj = re.search(
1763             r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1764         if mobj is not None:
1765             return self.url_result(mobj.group('url'))
1766
1767         # Look for NBC Sports VPlayer embeds
1768         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1769         if nbc_sports_url:
1770             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1771
1772         # Look for UDN embeds
1773         mobj = re.search(
1774             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage)
1775         if mobj is not None:
1776             return self.url_result(
1777                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1778
1779         # Look for Senate ISVP iframe
1780         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
1781         if senate_isvp_url:
1782             return self.url_result(senate_isvp_url, 'SenateISVP')
1783
1784         # Look for Dailymotion Cloud videos
1785         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
1786         if dmcloud_url:
1787             return self.url_result(dmcloud_url, 'DailymotionCloud')
1788
1789         # Look for OnionStudios embeds
1790         onionstudios_url = OnionStudiosIE._extract_url(webpage)
1791         if onionstudios_url:
1792             return self.url_result(onionstudios_url)
1793
1794         # Look for SnagFilms embeds
1795         snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage)
1796         if snagfilms_url:
1797             return self.url_result(snagfilms_url)
1798
1799         # Look for ScreenwaveMedia embeds
1800         mobj = re.search(ScreenwaveMediaIE.EMBED_PATTERN, webpage)
1801         if mobj is not None:
1802             return self.url_result(unescapeHTML(mobj.group('url')), 'ScreenwaveMedia')
1803
1804         # Look for AdobeTVVideo embeds
1805         mobj = re.search(
1806             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
1807             webpage)
1808         if mobj is not None:
1809             return self.url_result(
1810                 self._proto_relative_url(unescapeHTML(mobj.group(1))),
1811                 'AdobeTVVideo')
1812
1813         def check_video(vurl):
1814             if YoutubeIE.suitable(vurl):
1815                 return True
1816             vpath = compat_urlparse.urlparse(vurl).path
1817             vext = determine_ext(vpath)
1818             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1819
1820         def filter_video(urls):
1821             return list(filter(check_video, urls))
1822
1823         # Start with something easy: JW Player in SWFObject
1824         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1825         if not found:
1826             # Look for gorilla-vid style embedding
1827             found = filter_video(re.findall(r'''(?sx)
1828                 (?:
1829                     jw_plugins|
1830                     JWPlayerOptions|
1831                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1832                 )
1833                 .*?
1834                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1835         if not found:
1836             # Broaden the search a little bit
1837             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1838         if not found:
1839             # Broaden the findall a little bit: JWPlayer JS loader
1840             found = filter_video(re.findall(
1841                 r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1842         if not found:
1843             # Flow player
1844             found = filter_video(re.findall(r'''(?xs)
1845                 flowplayer\("[^"]+",\s*
1846                     \{[^}]+?\}\s*,
1847                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1848                         ["']?url["']?\s*:\s*["']([^"']+)["']
1849             ''', webpage))
1850         if not found:
1851             # Cinerama player
1852             found = re.findall(
1853                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1854         if not found:
1855             # Try to find twitter cards info
1856             found = filter_video(re.findall(
1857                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1858         if not found:
1859             # We look for Open Graph info:
1860             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1861             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1862             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1863             if m_video_type is not None:
1864                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1865         if not found:
1866             # HTML5 video
1867             found = re.findall(r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1868         if not found:
1869             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1870             found = re.search(
1871                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1872                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1873                 webpage)
1874             if not found:
1875                 # Look also in Refresh HTTP header
1876                 refresh_header = head_response.headers.get('Refresh')
1877                 if refresh_header:
1878                     # In python 2 response HTTP headers are bytestrings
1879                     if sys.version_info < (3, 0) and isinstance(refresh_header, str):
1880                         refresh_header = refresh_header.decode('iso-8859-1')
1881                     found = re.search(REDIRECT_REGEX, refresh_header)
1882             if found:
1883                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
1884                 self.report_following_redirect(new_url)
1885                 return {
1886                     '_type': 'url',
1887                     'url': new_url,
1888                 }
1889         if not found:
1890             raise UnsupportedError(url)
1891
1892         entries = []
1893         for video_url in found:
1894             video_url = video_url.replace('\\/', '/')
1895             video_url = compat_urlparse.urljoin(url, video_url)
1896             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
1897
1898             # Sometimes, jwplayer extraction will result in a YouTube URL
1899             if YoutubeIE.suitable(video_url):
1900                 entries.append(self.url_result(video_url, 'Youtube'))
1901                 continue
1902
1903             # here's a fun little line of code for you:
1904             video_id = os.path.splitext(video_id)[0]
1905
1906             entry_info_dict = {
1907                 'id': video_id,
1908                 'uploader': video_uploader,
1909                 'title': video_title,
1910                 'age_limit': age_limit,
1911             }
1912
1913             ext = determine_ext(video_url)
1914             if ext == 'smil':
1915                 entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
1916             elif ext == 'xspf':
1917                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
1918             elif ext == 'm3u8':
1919                 entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
1920             else:
1921                 entry_info_dict['url'] = video_url
1922
1923             entries.append(entry_info_dict)
1924
1925         if len(entries) == 1:
1926             return entries[0]
1927         else:
1928             for num, e in enumerate(entries, start=1):
1929                 # 'url' results don't have a title
1930                 if e.get('title') is not None:
1931                     e['title'] = '%s (%d)' % (e['title'], num)
1932             return {
1933                 '_type': 'playlist',
1934                 'entries': entries,
1935             }