_ 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 .svt import SVTIE
  48 from .pornhub import PornHubIE
  49 from .xhamster import XHamsterEmbedIE
  50 from .vimeo import VimeoIE
  51 from .dailymotion import DailymotionCloudIE
  52 from .onionstudios import OnionStudiosIE
  53 from .snagfilms import SnagFilmsEmbedIE
  54 from .screenwavemedia import ScreenwaveMediaIE
  55 from .mtv import MTVServicesEmbeddedIE
  56 from .pladform import PladformIE
  57 from .googledrive import GoogleDriveIE
  58 from .jwplatform import JWPlatformIE
  59
  60
  61 class GenericIE(InfoExtractor):
  62     IE_DESC = 'Generic downloader that works on some sites'
  63     _VALID_URL = r'.*'
  64     IE_NAME = 'generic'
  65     _TESTS = [
  66         # Direct link to a video
  67         {
  68             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  69             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  70             'info_dict': {
  71                 'id': 'trailer',
  72                 'ext': 'mp4',
  73                 'title': 'trailer',
  74                 'upload_date': '20100513',
  75             }
  76         },
  77         # Direct link to media delivered compressed (until Accept-Encoding is *)
  78         {
  79             'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
  80             'md5': '128c42e68b13950268b648275386fc74',
  81             'info_dict': {
  82                 'id': 'FictionJunction-Parallel_Hearts',
  83                 'ext': 'flac',
  84                 'title': 'FictionJunction-Parallel_Hearts',
  85                 'upload_date': '20140522',
  86             },
  87             'expected_warnings': [
  88                 'URL could be a direct video link, returning it as such.'
  89             ]
  90         },
  91         # Direct download with broken HEAD
  92         {
  93             'url': 'http://ai-radio.org:8000/radio.opus',
  94             'info_dict': {
  95                 'id': 'radio',
  96                 'ext': 'opus',
  97                 'title': 'radio',
  98             },
  99             'params': {
 100                 'skip_download': True,  # infinite live stream
 101             },
 102             'expected_warnings': [
 103                 r'501.*Not Implemented'
 104             ],
 105         },
 106         # Direct link with incorrect MIME type
 107         {
 108             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 109             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 110             'info_dict': {
 111                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 112                 'id': '5_Lennart_Poettering_-_Systemd',
 113                 'ext': 'webm',
 114                 'title': '5_Lennart_Poettering_-_Systemd',
 115                 'upload_date': '20141120',
 116             },
 117             'expected_warnings': [
 118                 'URL could be a direct video link, returning it as such.'
 119             ]
 120         },
 121         # RSS feed
 122         {
 123             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 124             'info_dict': {
 125                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 126                 'title': 'Zero Punctuation',
 127                 'description': 're:.*groundbreaking video review series.*'
 128             },
 129             'playlist_mincount': 11,
 130         },
 131         # RSS feed with enclosure
 132         {
 133             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 134             'info_dict': {
 135                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 136                 'ext': 'm4v',
 137                 'upload_date': '20150228',
 138                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 139             }
 140         },
 141         # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
 142         {
 143             'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
 144             'info_dict': {
 145                 'id': 'smil',
 146                 'ext': 'mp4',
 147                 'title': 'Automatics, robotics and biocybernetics',
 148                 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
 149                 'upload_date': '20130627',
 150                 'formats': 'mincount:16',
 151                 'subtitles': 'mincount:1',
 152             },
 153             'params': {
 154                 'force_generic_extractor': True,
 155                 'skip_download': True,
 156             },
 157         },
 158         # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html
 159         {
 160             'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil',
 161             'info_dict': {
 162                 'id': 'hds',
 163                 'ext': 'flv',
 164                 'title': 'hds',
 165                 'formats': 'mincount:1',
 166             },
 167             'params': {
 168                 'skip_download': True,
 169             },
 170         },
 171         # SMIL from https://www.restudy.dk/video/play/id/1637
 172         {
 173             'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml',
 174             'info_dict': {
 175                 'id': 'video_1637',
 176                 'ext': 'flv',
 177                 'title': 'video_1637',
 178                 'formats': 'mincount:3',
 179             },
 180             'params': {
 181                 'skip_download': True,
 182             },
 183         },
 184         # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm
 185         {
 186             'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil',
 187             'info_dict': {
 188                 'id': 'smil-service',
 189                 'ext': 'flv',
 190                 'title': 'smil-service',
 191                 'formats': 'mincount:1',
 192             },
 193             'params': {
 194                 'skip_download': True,
 195             },
 196         },
 197         # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370
 198         {
 199             'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil',
 200             'info_dict': {
 201                 'id': '4719370',
 202                 'ext': 'mp4',
 203                 'title': '571de1fd-47bc-48db-abf9-238872a58d1f',
 204                 'formats': 'mincount:3',
 205             },
 206             'params': {
 207                 'skip_download': True,
 208             },
 209         },
 210         # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html
 211         {
 212             'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf',
 213             'info_dict': {
 214                 'id': 'mZlp2ctYIUEB',
 215                 'ext': 'mp4',
 216                 'title': 'Tikibad ontruimd wegens brand',
 217                 'description': 'md5:05ca046ff47b931f9b04855015e163a4',
 218                 'thumbnail': 're:^https?://.*\.jpg$',
 219                 'duration': 33,
 220             },
 221             'params': {
 222                 'skip_download': True,
 223             },
 224         },
 225         # google redirect
 226         {
 227             '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',
 228             'info_dict': {
 229                 'id': 'cmQHVoWB5FY',
 230                 'ext': 'mp4',
 231                 'upload_date': '20130224',
 232                 'uploader_id': 'TheVerge',
 233                 'description': 're:^Chris Ziegler takes a look at the\.*',
 234                 'uploader': 'The Verge',
 235                 'title': 'First Firefox OS phones side-by-side',
 236             },
 237             'params': {
 238                 'skip_download': False,
 239             }
 240         },
 241         {
 242             # redirect in Refresh HTTP header
 243             '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',
 244             'info_dict': {
 245                 'id': 'pO8h3EaFRdo',
 246                 'ext': 'mp4',
 247                 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
 248                 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5',
 249                 'upload_date': '20150917',
 250                 'uploader_id': 'brtvofficial',
 251                 'uploader': 'Boiler Room',
 252             },
 253             'params': {
 254                 'skip_download': False,
 255             },
 256         },
 257         {
 258             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
 259             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
 260             'info_dict': {
 261                 'id': '13601338388002',
 262                 'ext': 'mp4',
 263                 'uploader': 'www.hodiho.fr',
 264                 'title': 'R\u00e9gis plante sa Jeep',
 265             }
 266         },
 267         # bandcamp page with custom domain
 268         {
 269             'add_ie': ['Bandcamp'],
 270             'url': 'http://bronyrock.com/track/the-pony-mash',
 271             'info_dict': {
 272                 'id': '3235767654',
 273                 'ext': 'mp3',
 274                 'title': 'The Pony Mash',
 275                 'uploader': 'M_Pallante',
 276             },
 277             'skip': 'There is a limit of 200 free downloads / month for the test song',
 278         },
 279         # embedded brightcove video
 280         # it also tests brightcove videos that need to set the 'Referer' in the
 281         # http requests
 282         {
 283             'add_ie': ['BrightcoveLegacy'],
 284             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
 285             'info_dict': {
 286                 'id': '2765128793001',
 287                 'ext': 'mp4',
 288                 'title': 'Le cours de bourse : l’analyse technique',
 289                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
 290                 'uploader': 'BFM BUSINESS',
 291             },
 292             'params': {
 293                 'skip_download': True,
 294             },
 295         },
 296         {
 297             # https://github.com/rg3/youtube-dl/issues/2253
 298             'url': 'http://bcove.me/i6nfkrc3',
 299             'md5': '0ba9446db037002366bab3b3eb30c88c',
 300             'info_dict': {
 301                 'id': '3101154703001',
 302                 'ext': 'mp4',
 303                 'title': 'Still no power',
 304                 'uploader': 'thestar.com',
 305                 '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.',
 306             },
 307             'add_ie': ['BrightcoveLegacy'],
 308         },
 309         {
 310             'url': 'http://www.championat.com/video/football/v/87/87499.html',
 311             'md5': 'fb973ecf6e4a78a67453647444222983',
 312             'info_dict': {
 313                 'id': '3414141473001',
 314                 'ext': 'mp4',
 315                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 316                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 317                 'uploader': 'Championat',
 318             },
 319         },
 320         {
 321             # https://github.com/rg3/youtube-dl/issues/3541
 322             'add_ie': ['BrightcoveLegacy'],
 323             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 324             'info_dict': {
 325                 'id': '3866516442001',
 326                 'ext': 'mp4',
 327                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 328                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 329                 'uploader': 'SBS Broadcasting',
 330             },
 331             'skip': 'Restricted to Netherlands',
 332             'params': {
 333                 'skip_download': True,  # m3u8 download
 334             },
 335         },
 336         # ooyala video
 337         {
 338             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 339             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 340             'info_dict': {
 341                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 342                 'ext': 'mp4',
 343                 'title': '2cc213299525360.mov',  # that's what we get
 344                 'duration': 238.231,
 345             },
 346             'add_ie': ['Ooyala'],
 347         },
 348         {
 349             # ooyala video embedded with http://player.ooyala.com/iframe.js
 350             'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/',
 351             'info_dict': {
 352                 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB',
 353                 'ext': 'mp4',
 354                 'title': '"Steve Jobs: Man in the Machine" trailer',
 355                 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
 356                 'duration': 135.427,
 357             },
 358             'params': {
 359                 'skip_download': True,
 360             },
 361         },
 362         # multiple ooyala embeds on SBN network websites
 363         {
 364             'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 365             'info_dict': {
 366                 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 367                 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
 368             },
 369             'playlist_mincount': 3,
 370             'params': {
 371                 'skip_download': True,
 372             },
 373             'add_ie': ['Ooyala'],
 374         },
 375         # embed.ly video
 376         {
 377             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 378             'info_dict': {
 379                 'id': '9ODmcdjQcHQ',
 380                 'ext': 'mp4',
 381                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 382                 'upload_date': '20140225',
 383                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 384                 'uploader': 'Tested',
 385                 'uploader_id': 'testedcom',
 386             },
 387             # No need to test YoutubeIE here
 388             'params': {
 389                 'skip_download': True,
 390             },
 391         },
 392         # funnyordie embed
 393         {
 394             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 395             'info_dict': {
 396                 'id': '18e820ec3f',
 397                 'ext': 'mp4',
 398                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 399                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 400             },
 401         },
 402         # RUTV embed
 403         {
 404             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 405             'info_dict': {
 406                 'id': '776940',
 407                 'ext': 'mp4',
 408                 'title': 'Охотское море стало целиком российским',
 409                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 410             },
 411             'params': {
 412                 # m3u8 download
 413                 'skip_download': True,
 414             },
 415         },
 416         # TVC embed
 417         {
 418             '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/',
 419             'info_dict': {
 420                 'id': '55304',
 421                 'ext': 'mp4',
 422                 'title': 'Дошкольное воспитание',
 423             },
 424         },
 425         # SportBox embed
 426         {
 427             'url': 'http://www.vestifinance.ru/articles/25753',
 428             'info_dict': {
 429                 'id': '25753',
 430                 'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
 431             },
 432             'playlist': [{
 433                 'info_dict': {
 434                     'id': '370908',
 435                     'title': 'Госзаказ. День 3',
 436                     'ext': 'mp4',
 437                 }
 438             }, {
 439                 'info_dict': {
 440                     'id': '370905',
 441                     'title': 'Госзаказ. День 2',
 442                     'ext': 'mp4',
 443                 }
 444             }, {
 445                 'info_dict': {
 446                     'id': '370902',
 447                     'title': 'Госзаказ. День 1',
 448                     'ext': 'mp4',
 449                 }
 450             }],
 451             'params': {
 452                 # m3u8 download
 453                 'skip_download': True,
 454             },
 455         },
 456         # Myvi.ru embed
 457         {
 458             'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
 459             'info_dict': {
 460                 'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
 461                 'ext': 'mp4',
 462                 'title': 'Ужастики, русский трейлер (2015)',
 463                 'thumbnail': 're:^https?://.*\.jpg$',
 464                 'duration': 153,
 465             }
 466         },
 467         # XHamster embed
 468         {
 469             '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',
 470             'info_dict': {
 471                 'id': 'showthread',
 472                 'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
 473             },
 474             'playlist_mincount': 7,
 475         },
 476         # Embedded TED video
 477         {
 478             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 479             'md5': '65fdff94098e4a607385a60c5177c638',
 480             'info_dict': {
 481                 'id': '1969',
 482                 'ext': 'mp4',
 483                 'title': 'Hidden miracles of the natural world',
 484                 'uploader': 'Louie Schwartzberg',
 485                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 486             }
 487         },
 488         # Embeded Ustream video
 489         {
 490             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 491             'md5': '27b99cdb639c9b12a79bca876a073417',
 492             'info_dict': {
 493                 'id': '45734260',
 494                 'ext': 'flv',
 495                 'uploader': 'AU SPA:  The NSA and Privacy',
 496                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 497             }
 498         },
 499         # nowvideo embed hidden behind percent encoding
 500         {
 501             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 502             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 503             'info_dict': {
 504                 'id': '06e53103ca9aa',
 505                 'ext': 'flv',
 506                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 507                 'description': 'No description',
 508             },
 509         },
 510         # arte embed
 511         {
 512             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 513             'md5': '7653032cbb25bf6c80d80f217055fa43',
 514             'info_dict': {
 515                 'id': '048195-004_PLUS7-F',
 516                 'ext': 'flv',
 517                 'title': 'X:enius',
 518                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 519                 'upload_date': '20140320',
 520             },
 521             'params': {
 522                 'skip_download': 'Requires rtmpdump'
 523             }
 524         },
 525         # francetv embed
 526         {
 527             'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
 528             'info_dict': {
 529                 'id': 'EV_30231',
 530                 'ext': 'mp4',
 531                 'title': 'Alcaline, le concert avec Calogero',
 532                 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
 533                 'upload_date': '20150226',
 534                 'timestamp': 1424989860,
 535                 'duration': 5400,
 536             },
 537             'params': {
 538                 # m3u8 downloads
 539                 'skip_download': True,
 540             },
 541             'expected_warnings': [
 542                 'Forbidden'
 543             ]
 544         },
 545         # Condé Nast embed
 546         {
 547             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 548             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 549             'info_dict': {
 550                 'id': '53501be369702d3275860000',
 551                 'ext': 'mp4',
 552                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 553             }
 554         },
 555         # Dailymotion embed
 556         {
 557             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 558             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 559             'info_dict': {
 560                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 561                 'ext': 'mp4',
 562                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 563                 'uploader': 'Spi0n',
 564             },
 565             'add_ie': ['Dailymotion'],
 566         },
 567         # YouTube embed
 568         {
 569             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 570             'info_dict': {
 571                 'id': 'FXRb4ykk4S0',
 572                 'ext': 'mp4',
 573                 'title': 'The NBL Auction 2014',
 574                 'uploader': 'BADMINTON England',
 575                 'uploader_id': 'BADMINTONEvents',
 576                 'upload_date': '20140603',
 577                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 578             },
 579             'add_ie': ['Youtube'],
 580             'params': {
 581                 'skip_download': True,
 582             }
 583         },
 584         # MTVSercices embed
 585         {
 586             'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
 587             'md5': '35727f82f58c76d996fc188f9755b0d5',
 588             'info_dict': {
 589                 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
 590                 'ext': 'mp4',
 591                 'title': 'Review',
 592                 'description': 'Mario\'s life in the fast lane has never looked so good.',
 593             },
 594         },
 595         # YouTube embed via <data-embed-url="">
 596         {
 597             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 598             'info_dict': {
 599                 'id': '4vAffPZIT44',
 600                 'ext': 'mp4',
 601                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 602                 'uploader': 'Gameloft',
 603                 'uploader_id': 'gameloft',
 604                 'upload_date': '20140828',
 605                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 606             },
 607             'params': {
 608                 'skip_download': True,
 609             }
 610         },
 611         # Camtasia studio
 612         {
 613             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 614             'playlist': [{
 615                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 616                 'info_dict': {
 617                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 618                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 619                     'ext': 'flv',
 620                     'duration': 2235.90,
 621                 }
 622             }, {
 623                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 624                 'info_dict': {
 625                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 626                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 627                     'ext': 'flv',
 628                     'duration': 2235.93,
 629                 }
 630             }],
 631             'info_dict': {
 632                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 633             }
 634         },
 635         # Flowplayer
 636         {
 637             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 638             'md5': '9d65602bf31c6e20014319c7d07fba27',
 639             'info_dict': {
 640                 'id': '5123ea6d5e5a7',
 641                 'ext': 'mp4',
 642                 'age_limit': 18,
 643                 'uploader': 'www.handjobhub.com',
 644                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 645             }
 646         },
 647         # Multiple brightcove videos
 648         # https://github.com/rg3/youtube-dl/issues/2283
 649         {
 650             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 651             'info_dict': {
 652                 'id': 'always-never',
 653                 'title': 'Always / Never - The New Yorker',
 654             },
 655             'playlist_count': 3,
 656             'params': {
 657                 'extract_flat': False,
 658                 'skip_download': True,
 659             }
 660         },
 661         # MLB embed
 662         {
 663             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 664             'md5': '96f09a37e44da40dd083e12d9a683327',
 665             'info_dict': {
 666                 'id': '33322633',
 667                 'ext': 'mp4',
 668                 'title': 'Ump changes call to ball',
 669                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 670                 'duration': 48,
 671                 'timestamp': 1401537900,
 672                 'upload_date': '20140531',
 673                 'thumbnail': 're:^https?://.*\.jpg$',
 674             },
 675         },
 676         # Wistia embed
 677         {
 678             'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 679             'md5': '8788b683c777a5cf25621eaf286d0c23',
 680             'info_dict': {
 681                 'id': '1cfaf6b7ea',
 682                 'ext': 'mov',
 683                 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
 684                 'duration': 643.0,
 685                 'filesize': 182808282,
 686                 'uploader': 'education-portal.com',
 687             },
 688         },
 689         {
 690             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 691             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 692             'info_dict': {
 693                 'id': 'uxjb0lwrcz',
 694                 'ext': 'mp4',
 695                 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
 696                 'duration': 1715.0,
 697                 'uploader': 'thoughtworks.wistia.com',
 698             },
 699         },
 700         # Soundcloud embed
 701         {
 702             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 703             'info_dict': {
 704                 'id': '174391317',
 705                 'ext': 'mp3',
 706                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 707                 'uploader': 'Sophos Security',
 708                 'title': 'Chet Chat 171 - Oct 29, 2014',
 709                 'upload_date': '20141029',
 710             }
 711         },
 712         # Livestream embed
 713         {
 714             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 715             'info_dict': {
 716                 'id': '67864563',
 717                 'ext': 'flv',
 718                 'upload_date': '20141112',
 719                 'title': 'Rosetta #CometLanding webcast HL 10',
 720             }
 721         },
 722         # LazyYT
 723         {
 724             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
 725             'info_dict': {
 726                 'id': '1986',
 727                 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
 728             },
 729             'playlist_mincount': 2,
 730         },
 731         # Cinchcast embed
 732         {
 733             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
 734             'info_dict': {
 735                 'id': '7141703',
 736                 'ext': 'mp3',
 737                 'upload_date': '20141126',
 738                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
 739             }
 740         },
 741         # Cinerama player
 742         {
 743             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
 744             'info_dict': {
 745                 'id': '730m_DandD_1901_512k',
 746                 'ext': 'mp4',
 747                 'uploader': 'www.abc.net.au',
 748                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
 749             }
 750         },
 751         # embedded viddler video
 752         {
 753             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
 754             'info_dict': {
 755                 'id': '4d03aad9',
 756                 'ext': 'mp4',
 757                 'uploader': 'deadspin',
 758                 'title': 'WALL-TO-GORTAT',
 759                 'timestamp': 1422285291,
 760                 'upload_date': '20150126',
 761             },
 762             'add_ie': ['Viddler'],
 763         },
 764         # Libsyn embed
 765         {
 766             'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
 767             'info_dict': {
 768                 'id': '3377616',
 769                 'ext': 'mp3',
 770                 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
 771                 'description': 'md5:601cb790edd05908957dae8aaa866465',
 772                 'upload_date': '20150220',
 773             },
 774         },
 775         # jwplayer YouTube
 776         {
 777             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
 778             'info_dict': {
 779                 'id': 'Mrj4DVp2zeA',
 780                 'ext': 'mp4',
 781                 'upload_date': '20150212',
 782                 'uploader': 'The National Archives UK',
 783                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
 784                 'uploader_id': 'NationalArchives08',
 785                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
 786             },
 787         },
 788         # rtl.nl embed
 789         {
 790             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
 791             'playlist_mincount': 5,
 792             'info_dict': {
 793                 'id': 'aanslagen-kopenhagen',
 794                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
 795             }
 796         },
 797         # Zapiks embed
 798         {
 799             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
 800             'info_dict': {
 801                 'id': '118046',
 802                 'ext': 'mp4',
 803                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
 804             }
 805         },
 806         # Kaltura embed
 807         {
 808             'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
 809             'info_dict': {
 810                 'id': '1_eergr3h1',
 811                 'ext': 'mp4',
 812                 'upload_date': '20150226',
 813                 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
 814                 'timestamp': int,
 815                 'title': 'John Carlson Postgame 2/25/15',
 816             },
 817         },
 818         # Kaltura embed (different embed code)
 819         {
 820             'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
 821             'info_dict': {
 822                 'id': '1_a52wc67y',
 823                 'ext': 'flv',
 824                 'upload_date': '20150127',
 825                 'uploader_id': 'PremierMedia',
 826                 'timestamp': int,
 827                 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
 828             },
 829         },
 830         # Kaltura embed protected with referrer
 831         {
 832             'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero',
 833             'info_dict': {
 834                 'id': '1_g4fbemnq',
 835                 'ext': 'mp4',
 836                 'title': 'Violetta - Achter De Schermen - Ruggero',
 837                 'description': 'Achter de schermen met Ruggero',
 838                 'timestamp': 1435133761,
 839                 'upload_date': '20150624',
 840                 'uploader_id': 'echojecka',
 841             },
 842         },
 843         # Eagle.Platform embed (generic URL)
 844         {
 845             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
 846             'info_dict': {
 847                 'id': '227304',
 848                 'ext': 'mp4',
 849                 'title': 'Навальный вышел на свободу',
 850                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
 851                 'thumbnail': 're:^https?://.*\.jpg$',
 852                 'duration': 87,
 853                 'view_count': int,
 854                 'age_limit': 0,
 855             },
 856         },
 857         # ClipYou (Eagle.Platform) embed (custom URL)
 858         {
 859             'url': 'http://muz-tv.ru/play/7129/',
 860             'info_dict': {
 861                 'id': '12820',
 862                 'ext': 'mp4',
 863                 'title': "'O Sole Mio",
 864                 'thumbnail': 're:^https?://.*\.jpg$',
 865                 'duration': 216,
 866                 'view_count': int,
 867             },
 868         },
 869         # Pladform embed
 870         {
 871             'url': 'http://muz-tv.ru/kinozal/view/7400/',
 872             'info_dict': {
 873                 'id': '100183293',
 874                 'ext': 'mp4',
 875                 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
 876                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
 877                 'thumbnail': 're:^https?://.*\.jpg$',
 878                 'duration': 694,
 879                 'age_limit': 0,
 880             },
 881         },
 882         # Playwire embed
 883         {
 884             'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
 885             'info_dict': {
 886                 'id': '3519514',
 887                 'ext': 'mp4',
 888                 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
 889                 'thumbnail': 're:^https?://.*\.png$',
 890                 'duration': 45.115,
 891             },
 892         },
 893         # 5min embed
 894         {
 895             'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
 896             'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
 897             'info_dict': {
 898                 'id': '518726732',
 899                 'ext': 'mp4',
 900                 'title': 'Facebook Creates "On This Day" | Crunch Report',
 901             },
 902         },
 903         # SVT embed
 904         {
 905             'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 906             'info_dict': {
 907                 'id': '2900353',
 908                 'ext': 'flv',
 909                 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
 910                 'duration': 27,
 911                 'age_limit': 0,
 912             },
 913         },
 914         # Crooks and Liars embed
 915         {
 916             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
 917             'info_dict': {
 918                 'id': '8RUoRhRi',
 919                 'ext': 'mp4',
 920                 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
 921                 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
 922                 'timestamp': 1428207000,
 923                 'upload_date': '20150405',
 924                 'uploader': 'Heather',
 925             },
 926         },
 927         # Crooks and Liars external embed
 928         {
 929             'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
 930             'info_dict': {
 931                 'id': 'MTE3MjUtMzQ2MzA',
 932                 'ext': 'mp4',
 933                 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
 934                 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
 935                 'timestamp': 1265032391,
 936                 'upload_date': '20100201',
 937                 'uploader': 'Heather',
 938             },
 939         },
 940         # NBC Sports vplayer embed
 941         {
 942             'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
 943             'info_dict': {
 944                 'id': 'ln7x1qSThw4k',
 945                 'ext': 'flv',
 946                 'title': "PFT Live: New leader in the 'new-look' defense",
 947                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
 948             },
 949         },
 950         # UDN embed
 951         {
 952             'url': 'http://www.udn.com/news/story/7314/822787',
 953             'md5': 'fd2060e988c326991037b9aff9df21a6',
 954             'info_dict': {
 955                 'id': '300346',
 956                 'ext': 'mp4',
 957                 'title': '中一中男師變性 全校師生力挺',
 958                 'thumbnail': 're:^https?://.*\.jpg$',
 959             }
 960         },
 961         # Ooyala embed
 962         {
 963             'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
 964             'info_dict': {
 965                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
 966                 'ext': 'mp4',
 967                 'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.',
 968                 'title': 'This is what separates the Excel masters from the wannabes',
 969                 'duration': 191.933,
 970             },
 971             'params': {
 972                 # m3u8 downloads
 973                 'skip_download': True,
 974             }
 975         },
 976         # Contains a SMIL manifest
 977         {
 978             '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',
 979             'info_dict': {
 980                 'id': 'file',
 981                 'ext': 'flv',
 982                 'title': '+ Football: Lottery Champions League Europe',
 983                 'uploader': 'www.telewebion.com',
 984             },
 985             'params': {
 986                 # rtmpe downloads
 987                 'skip_download': True,
 988             }
 989         },
 990         # Brightcove URL in single quotes
 991         {
 992             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
 993             'md5': '4ae374f1f8b91c889c4b9203c8c752af',
 994             'info_dict': {
 995                 'id': '4255764656001',
 996                 'ext': 'mp4',
 997                 'title': 'SN Presents: Russell Martin, World Citizen',
 998                 '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.',
 999                 'uploader': 'Rogers Sportsnet',
1000             },
1001         },
1002         # Dailymotion Cloud video
1003         {
1004             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
1005             'md5': '49444254273501a64675a7e68c502681',
1006             'info_dict': {
1007                 'id': '5585de919473990de4bee11b',
1008                 'ext': 'mp4',
1009                 'title': 'Le débat',
1010                 'thumbnail': 're:^https?://.*\.jpe?g$',
1011             }
1012         },
1013         # OnionStudios embed
1014         {
1015             'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
1016             'info_dict': {
1017                 'id': '2855',
1018                 'ext': 'mp4',
1019                 'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
1020                 'thumbnail': 're:^https?://.*\.jpe?g$',
1021                 'uploader': 'ClickHole',
1022                 'uploader_id': 'clickhole',
1023             }
1024         },
1025         # SnagFilms embed
1026         {
1027             'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
1028             'info_dict': {
1029                 'id': '74849a00-85a9-11e1-9660-123139220831',
1030                 'ext': 'mp4',
1031                 'title': '#whilewewatch',
1032             }
1033         },
1034         # AdobeTVVideo embed
1035         {
1036             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
1037             'md5': '43662b577c018ad707a63766462b1e87',
1038             'info_dict': {
1039                 'id': '2456',
1040                 'ext': 'mp4',
1041                 'title': 'New experience with Acrobat DC',
1042                 'description': 'New experience with Acrobat DC',
1043                 'duration': 248.667,
1044             },
1045         },
1046         # ScreenwaveMedia embed
1047         {
1048             'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1',
1049             'md5': '24ace5baba0d35d55c6810b51f34e9e0',
1050             'info_dict': {
1051                 'id': 'cinemasnob-55d26273809dd',
1052                 'ext': 'mp4',
1053                 'title': 'cinemasnob',
1054             },
1055         },
1056         # BrightcoveInPageEmbed embed
1057         {
1058             'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
1059             'info_dict': {
1060                 'id': '4238694884001',
1061                 'ext': 'flv',
1062                 'title': 'Tabletop: Dread, Last Thoughts',
1063                 'description': 'Tabletop: Dread, Last Thoughts',
1064                 'duration': 51690,
1065             },
1066         },
1067         # JWPlayer with M3U8
1068         {
1069             'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video',
1070             'info_dict': {
1071                 'id': 'playlist',
1072                 'ext': 'mp4',
1073                 'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ',
1074                 'uploader': 'ren.tv',
1075             },
1076             'params': {
1077                 # m3u8 downloads
1078                 'skip_download': True,
1079             }
1080         }
1081     ]
1082
1083     def report_following_redirect(self, new_url):
1084         """Report information extraction."""
1085         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
1086
1087     def _extract_rss(self, url, video_id, doc):
1088         playlist_title = doc.find('./channel/title').text
1089         playlist_desc_el = doc.find('./channel/description')
1090         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
1091
1092         entries = []
1093         for it in doc.findall('./channel/item'):
1094             next_url = xpath_text(it, 'link', fatal=False)
1095             if not next_url:
1096                 enclosure_nodes = it.findall('./enclosure')
1097                 for e in enclosure_nodes:
1098                     next_url = e.attrib.get('url')
1099                     if next_url:
1100                         break
1101
1102             if not next_url:
1103                 continue
1104
1105             entries.append({
1106                 '_type': 'url',
1107                 'url': next_url,
1108                 'title': it.find('title').text,
1109             })
1110
1111         return {
1112             '_type': 'playlist',
1113             'id': url,
1114             'title': playlist_title,
1115             'description': playlist_desc,
1116             'entries': entries,
1117         }
1118
1119     def _extract_camtasia(self, url, video_id, webpage):
1120         """ Returns None if no camtasia video can be found. """
1121
1122         camtasia_cfg = self._search_regex(
1123             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
1124             webpage, 'camtasia configuration file', default=None)
1125         if camtasia_cfg is None:
1126             return None
1127
1128         title = self._html_search_meta('DC.title', webpage, fatal=True)
1129
1130         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
1131         camtasia_cfg = self._download_xml(
1132             camtasia_url, video_id,
1133             note='Downloading camtasia configuration',
1134             errnote='Failed to download camtasia configuration')
1135         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
1136
1137         entries = []
1138         for n in fileset_node.getchildren():
1139             url_n = n.find('./uri')
1140             if url_n is None:
1141                 continue
1142
1143             entries.append({
1144                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
1145                 'title': '%s - %s' % (title, n.tag),
1146                 'url': compat_urlparse.urljoin(url, url_n.text),
1147                 'duration': float_or_none(n.find('./duration').text),
1148             })
1149
1150         return {
1151             '_type': 'playlist',
1152             'entries': entries,
1153             'title': title,
1154         }
1155
1156     def _real_extract(self, url):
1157         if url.startswith('//'):
1158             return {
1159                 '_type': 'url',
1160                 'url': self.http_scheme() + url,
1161             }
1162
1163         parsed_url = compat_urlparse.urlparse(url)
1164         if not parsed_url.scheme:
1165             default_search = self._downloader.params.get('default_search')
1166             if default_search is None:
1167                 default_search = 'fixup_error'
1168
1169             if default_search in ('auto', 'auto_warning', 'fixup_error'):
1170                 if '/' in url:
1171                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
1172                     return self.url_result('http://' + url)
1173                 elif default_search != 'fixup_error':
1174                     if default_search == 'auto_warning':
1175                         if re.match(r'^(?:url|URL)$', url):
1176                             raise ExtractorError(
1177                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
1178                                 expected=True)
1179                         else:
1180                             self._downloader.report_warning(
1181                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
1182                     return self.url_result('ytsearch:' + url)
1183
1184             if default_search in ('error', 'fixup_error'):
1185                 raise ExtractorError(
1186                     '%r is not a valid URL. '
1187                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
1188                     % (url, url), expected=True)
1189             else:
1190                 if ':' not in default_search:
1191                     default_search += ':'
1192                 return self.url_result(default_search + url)
1193
1194         url, smuggled_data = unsmuggle_url(url)
1195         force_videoid = None
1196         is_intentional = smuggled_data and smuggled_data.get('to_generic')
1197         if smuggled_data and 'force_videoid' in smuggled_data:
1198             force_videoid = smuggled_data['force_videoid']
1199             video_id = force_videoid
1200         else:
1201             video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
1202
1203         self.to_screen('%s: Requesting header' % video_id)
1204
1205         head_req = HEADRequest(url)
1206         head_response = self._request_webpage(
1207             head_req, video_id,
1208             note=False, errnote='Could not send HEAD request to %s' % url,
1209             fatal=False)
1210
1211         if head_response is not False:
1212             # Check for redirect
1213             new_url = head_response.geturl()
1214             if url != new_url:
1215                 self.report_following_redirect(new_url)
1216                 if force_videoid:
1217                     new_url = smuggle_url(
1218                         new_url, {'force_videoid': force_videoid})
1219                 return self.url_result(new_url)
1220
1221         full_response = None
1222         if head_response is False:
1223             request = sanitized_Request(url)
1224             request.add_header('Accept-Encoding', '*')
1225             full_response = self._request_webpage(request, video_id)
1226             head_response = full_response
1227
1228         # Check for direct link to a video
1229         content_type = head_response.headers.get('Content-Type', '')
1230         m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
1231         if m:
1232             upload_date = unified_strdate(
1233                 head_response.headers.get('Last-Modified'))
1234             return {
1235                 'id': video_id,
1236                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1237                 'direct': True,
1238                 'formats': [{
1239                     'format_id': m.group('format_id'),
1240                     'url': url,
1241                     'vcodec': 'none' if m.group('type') == 'audio' else None
1242                 }],
1243                 'upload_date': upload_date,
1244             }
1245
1246         if not self._downloader.params.get('test', False) and not is_intentional:
1247             force = self._downloader.params.get('force_generic_extractor', False)
1248             self._downloader.report_warning(
1249                 '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
1250
1251         if not full_response:
1252             request = sanitized_Request(url)
1253             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
1254             # making it impossible to download only chunk of the file (yet we need only 512kB to
1255             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
1256             # that will always result in downloading the whole file that is not desirable.
1257             # Therefore for extraction pass we have to override Accept-Encoding to any in order
1258             # to accept raw bytes and being able to download only a chunk.
1259             # It may probably better to solve this by checking Content-Type for application/octet-stream
1260             # after HEAD request finishes, but not sure if we can rely on this.
1261             request.add_header('Accept-Encoding', '*')
1262             full_response = self._request_webpage(request, video_id)
1263
1264         # Maybe it's a direct link to a video?
1265         # Be careful not to download the whole thing!
1266         first_bytes = full_response.read(512)
1267         if not is_html(first_bytes):
1268             self._downloader.report_warning(
1269                 'URL could be a direct video link, returning it as such.')
1270             upload_date = unified_strdate(
1271                 head_response.headers.get('Last-Modified'))
1272             return {
1273                 'id': video_id,
1274                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1275                 'direct': True,
1276                 'url': url,
1277                 'upload_date': upload_date,
1278             }
1279
1280         webpage = self._webpage_read_content(
1281             full_response, url, video_id, prefix=first_bytes)
1282
1283         self.report_extraction(video_id)
1284
1285         # Is it an RSS feed, a SMIL file or a XSPF playlist?
1286         try:
1287             doc = compat_etree_fromstring(webpage.encode('utf-8'))
1288             if doc.tag == 'rss':
1289                 return self._extract_rss(url, video_id, doc)
1290             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
1291                 return self._parse_smil(doc, url, video_id)
1292             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
1293                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
1294         except compat_xml_parse_error:
1295             pass
1296
1297         # Is it a Camtasia project?
1298         camtasia_res = self._extract_camtasia(url, video_id, webpage)
1299         if camtasia_res is not None:
1300             return camtasia_res
1301
1302         # Sometimes embedded video player is hidden behind percent encoding
1303         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1304         # Unescaping the whole page allows to handle those cases in a generic way
1305         webpage = compat_urllib_parse_unquote(webpage)
1306
1307         # it's tempting to parse this further, but you would
1308         # have to take into account all the variations like
1309         #   Video Title - Site Name
1310         #   Site Name | Video Title
1311         #   Video Title - Tagline | Site Name
1312         # and so on and so forth; it's just not practical
1313         video_title = self._html_search_regex(
1314             r'(?s)<title>(.*?)</title>', webpage, 'video title',
1315             default='video')
1316
1317         # Try to detect age limit automatically
1318         age_limit = self._rta_search(webpage)
1319         # And then there are the jokers who advertise that they use RTA,
1320         # but actually don't.
1321         AGE_LIMIT_MARKERS = [
1322             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1323         ]
1324         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1325             age_limit = 18
1326
1327         # video uploader is domain name
1328         video_uploader = self._search_regex(
1329             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1330
1331         # Helper method
1332         def _playlist_from_matches(matches, getter=None, ie=None):
1333             urlrs = orderedSet(
1334                 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1335                 for m in matches)
1336             return self.playlist_result(
1337                 urlrs, playlist_id=video_id, playlist_title=video_title)
1338
1339         # Look for Brightcove Legacy Studio embeds
1340         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
1341         if bc_urls:
1342             self.to_screen('Brightcove video detected.')
1343             entries = [{
1344                 '_type': 'url',
1345                 'url': smuggle_url(bc_url, {'Referer': url}),
1346                 'ie_key': 'BrightcoveLegacy'
1347             } for bc_url in bc_urls]
1348
1349             return {
1350                 '_type': 'playlist',
1351                 'title': video_title,
1352                 'id': video_id,
1353                 'entries': entries,
1354             }
1355
1356         # Look for Brightcove New Studio embeds
1357         bc_urls = BrightcoveNewIE._extract_urls(webpage)
1358         if bc_urls:
1359             return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
1360
1361         # Look for embedded rtl.nl player
1362         matches = re.findall(
1363             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1364             webpage)
1365         if matches:
1366             return _playlist_from_matches(matches, ie='RtlNl')
1367
1368         vimeo_url = VimeoIE._extract_vimeo_url(url, webpage)
1369         if vimeo_url is not None:
1370             return self.url_result(vimeo_url)
1371
1372         vid_me_embed_url = self._search_regex(
1373             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
1374             webpage, 'vid.me embed', default=None)
1375         if vid_me_embed_url is not None:
1376             return self.url_result(vid_me_embed_url, 'Vidme')
1377
1378         # Look for embedded YouTube player
1379         matches = re.findall(r'''(?x)
1380             (?:
1381                 <iframe[^>]+?src=|
1382                 data-video-url=|
1383                 <embed[^>]+?src=|
1384                 embedSWF\(?:\s*|
1385                 new\s+SWFObject\(
1386             )
1387             (["\'])
1388                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1389                 (?:embed|v|p)/.+?)
1390             \1''', webpage)
1391         if matches:
1392             return _playlist_from_matches(
1393                 matches, lambda m: unescapeHTML(m[1]))
1394
1395         # Look for lazyYT YouTube embed
1396         matches = re.findall(
1397             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1398         if matches:
1399             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1400
1401         # Look for embedded Dailymotion player
1402         matches = re.findall(
1403             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
1404         if matches:
1405             return _playlist_from_matches(
1406                 matches, lambda m: unescapeHTML(m[1]))
1407
1408         # Look for embedded Dailymotion playlist player (#3822)
1409         m = re.search(
1410             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1411         if m:
1412             playlists = re.findall(
1413                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1414             if playlists:
1415                 return _playlist_from_matches(
1416                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1417
1418         # Look for embedded Wistia player
1419         match = re.search(
1420             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1421         if match:
1422             embed_url = self._proto_relative_url(
1423                 unescapeHTML(match.group('url')))
1424             return {
1425                 '_type': 'url_transparent',
1426                 'url': embed_url,
1427                 'ie_key': 'Wistia',
1428                 'uploader': video_uploader,
1429                 'title': video_title,
1430                 'id': video_id,
1431             }
1432
1433         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1434         if match:
1435             return {
1436                 '_type': 'url_transparent',
1437                 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1438                 'ie_key': 'Wistia',
1439                 'uploader': video_uploader,
1440                 'title': video_title,
1441                 'id': match.group('id')
1442             }
1443
1444         # Look for SVT player
1445         svt_url = SVTIE._extract_url(webpage)
1446         if svt_url:
1447             return self.url_result(svt_url, 'SVT')
1448
1449         # Look for embedded condenast player
1450         matches = re.findall(
1451             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1452             webpage)
1453         if matches:
1454             return {
1455                 '_type': 'playlist',
1456                 'entries': [{
1457                     '_type': 'url',
1458                     'ie_key': 'CondeNast',
1459                     'url': ma,
1460                 } for ma in matches],
1461                 'title': video_title,
1462                 'id': video_id,
1463             }
1464
1465         # Look for Bandcamp pages with custom domain
1466         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1467         if mobj is not None:
1468             burl = unescapeHTML(mobj.group(1))
1469             # Don't set the extractor because it can be a track url or an album
1470             return self.url_result(burl)
1471
1472         # Look for embedded Vevo player
1473         mobj = re.search(
1474             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1475         if mobj is not None:
1476             return self.url_result(mobj.group('url'))
1477
1478         # Look for embedded Viddler player
1479         mobj = re.search(
1480             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1481             webpage)
1482         if mobj is not None:
1483             return self.url_result(mobj.group('url'))
1484
1485         # Look for NYTimes player
1486         mobj = re.search(
1487             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1488             webpage)
1489         if mobj is not None:
1490             return self.url_result(mobj.group('url'))
1491
1492         # Look for Libsyn player
1493         mobj = re.search(
1494             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1495         if mobj is not None:
1496             return self.url_result(mobj.group('url'))
1497
1498         # Look for Ooyala videos
1499         mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1500                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1501                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1502                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1503         if mobj is not None:
1504             return OoyalaIE._build_url_result(smuggle_url(mobj.group('ec'), {'domain': url}))
1505
1506         # Look for multiple Ooyala embeds on SBN network websites
1507         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1508         if mobj is not None:
1509             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1510             if embeds:
1511                 return _playlist_from_matches(
1512                     embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
1513
1514         # Look for Aparat videos
1515         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1516         if mobj is not None:
1517             return self.url_result(mobj.group(1), 'Aparat')
1518
1519         # Look for MPORA videos
1520         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1521         if mobj is not None:
1522             return self.url_result(mobj.group(1), 'Mpora')
1523
1524         # Look for embedded NovaMov-based player
1525         mobj = re.search(
1526             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1527                     (?P<url>http://(?:(?:embed|www)\.)?
1528                         (?:novamov\.com|
1529                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
1530                            videoweed\.(?:es|com)|
1531                            movshare\.(?:net|sx|ag)|
1532                            divxstage\.(?:eu|net|ch|co|at|ag))
1533                         /embed\.php.+?)\1''', webpage)
1534         if mobj is not None:
1535             return self.url_result(mobj.group('url'))
1536
1537         # Look for embedded Facebook player
1538         mobj = re.search(
1539             r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1540         if mobj is not None:
1541             return self.url_result(mobj.group('url'), 'Facebook')
1542
1543         # Look for embedded VK player
1544         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1545         if mobj is not None:
1546             return self.url_result(mobj.group('url'), 'VK')
1547
1548         # Look for embedded ivi player
1549         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1550         if mobj is not None:
1551             return self.url_result(mobj.group('url'), 'Ivi')
1552
1553         # Look for embedded Huffington Post player
1554         mobj = re.search(
1555             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1556         if mobj is not None:
1557             return self.url_result(mobj.group('url'), 'HuffPost')
1558
1559         # Look for embed.ly
1560         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1561         if mobj is not None:
1562             return self.url_result(mobj.group('url'))
1563         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1564         if mobj is not None:
1565             return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
1566
1567         # Look for funnyordie embed
1568         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1569         if matches:
1570             return _playlist_from_matches(
1571                 matches, getter=unescapeHTML, ie='FunnyOrDie')
1572
1573         # Look for BBC iPlayer embed
1574         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1575         if matches:
1576             return _playlist_from_matches(matches, ie='BBCCoUk')
1577
1578         # Look for embedded RUTV player
1579         rutv_url = RUTVIE._extract_url(webpage)
1580         if rutv_url:
1581             return self.url_result(rutv_url, 'RUTV')
1582
1583         # Look for embedded TVC player
1584         tvc_url = TVCIE._extract_url(webpage)
1585         if tvc_url:
1586             return self.url_result(tvc_url, 'TVC')
1587
1588         # Look for embedded SportBox player
1589         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1590         if sportbox_urls:
1591             return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1592
1593         # Look for embedded PornHub player
1594         pornhub_url = PornHubIE._extract_url(webpage)
1595         if pornhub_url:
1596             return self.url_result(pornhub_url, 'PornHub')
1597
1598         # Look for embedded XHamster player
1599         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
1600         if xhamster_urls:
1601             return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
1602
1603         # Look for embedded Tvigle player
1604         mobj = re.search(
1605             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
1606         if mobj is not None:
1607             return self.url_result(mobj.group('url'), 'Tvigle')
1608
1609         # Look for embedded TED player
1610         mobj = re.search(
1611             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1612         if mobj is not None:
1613             return self.url_result(mobj.group('url'), 'TED')
1614
1615         # Look for embedded Ustream videos
1616         mobj = re.search(
1617             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1618         if mobj is not None:
1619             return self.url_result(mobj.group('url'), 'Ustream')
1620
1621         # Look for embedded arte.tv player
1622         mobj = re.search(
1623             r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1624             webpage)
1625         if mobj is not None:
1626             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1627
1628         # Look for embedded francetv player
1629         mobj = re.search(
1630             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
1631             webpage)
1632         if mobj is not None:
1633             return self.url_result(mobj.group('url'))
1634
1635         # Look for embedded smotri.com player
1636         smotri_url = SmotriIE._extract_url(webpage)
1637         if smotri_url:
1638             return self.url_result(smotri_url, 'Smotri')
1639
1640         # Look for embedded Myvi.ru player
1641         myvi_url = MyviIE._extract_url(webpage)
1642         if myvi_url:
1643             return self.url_result(myvi_url)
1644
1645         # Look for embeded soundcloud player
1646         mobj = re.search(
1647             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1648             webpage)
1649         if mobj is not None:
1650             url = unescapeHTML(mobj.group('url'))
1651             return self.url_result(url)
1652
1653         # Look for embedded vulture.com player
1654         mobj = re.search(
1655             r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1656             webpage)
1657         if mobj is not None:
1658             url = unescapeHTML(mobj.group('url'))
1659             return self.url_result(url, ie='Vulture')
1660
1661         # Look for embedded mtvservices player
1662         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
1663         if mtvservices_url:
1664             return self.url_result(mtvservices_url, ie='MTVServicesEmbedded')
1665
1666         # Look for embedded yahoo player
1667         mobj = re.search(
1668             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1669             webpage)
1670         if mobj is not None:
1671             return self.url_result(mobj.group('url'), 'Yahoo')
1672
1673         # Look for embedded sbs.com.au player
1674         mobj = re.search(
1675             r'''(?x)
1676             (?:
1677                 <meta\s+property="og:video"\s+content=|
1678                 <iframe[^>]+?src=
1679             )
1680             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1681             webpage)
1682         if mobj is not None:
1683             return self.url_result(mobj.group('url'), 'SBS')
1684
1685         # Look for embedded Cinchcast player
1686         mobj = re.search(
1687             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1688             webpage)
1689         if mobj is not None:
1690             return self.url_result(mobj.group('url'), 'Cinchcast')
1691
1692         mobj = re.search(
1693             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1694             webpage)
1695         if not mobj:
1696             mobj = re.search(
1697                 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
1698                 webpage)
1699         if mobj is not None:
1700             return self.url_result(mobj.group('url'), 'MLB')
1701
1702         mobj = re.search(
1703             r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1704             webpage)
1705         if mobj is not None:
1706             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1707
1708         mobj = re.search(
1709             r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1710             webpage)
1711         if mobj is not None:
1712             return self.url_result(mobj.group('url'), 'Livestream')
1713
1714         # Look for Zapiks embed
1715         mobj = re.search(
1716             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1717         if mobj is not None:
1718             return self.url_result(mobj.group('url'), 'Zapiks')
1719
1720         # Look for Kaltura embeds
1721         mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or
1722                 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))
1723         if mobj is not None:
1724             return self.url_result(smuggle_url(
1725                 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(),
1726                 {'source_url': url}), 'Kaltura')
1727
1728         # Look for Eagle.Platform embeds
1729         mobj = re.search(
1730             r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1731         if mobj is not None:
1732             return self.url_result(mobj.group('url'), 'EaglePlatform')
1733
1734         # Look for ClipYou (uses Eagle.Platform) embeds
1735         mobj = re.search(
1736             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1737         if mobj is not None:
1738             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1739
1740         # Look for Pladform embeds
1741         pladform_url = PladformIE._extract_url(webpage)
1742         if pladform_url:
1743             return self.url_result(pladform_url)
1744
1745         # Look for Playwire embeds
1746         mobj = re.search(
1747             r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
1748         if mobj is not None:
1749             return self.url_result(mobj.group('url'))
1750
1751         # Look for 5min embeds
1752         mobj = re.search(
1753             r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1754         if mobj is not None:
1755             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1756
1757         # Look for Crooks and Liars embeds
1758         mobj = re.search(
1759             r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1760         if mobj is not None:
1761             return self.url_result(mobj.group('url'))
1762
1763         # Look for NBC Sports VPlayer embeds
1764         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1765         if nbc_sports_url:
1766             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1767
1768         # Look for Google Drive embeds
1769         google_drive_url = GoogleDriveIE._extract_url(webpage)
1770         if google_drive_url:
1771             return self.url_result(google_drive_url, 'GoogleDrive')
1772
1773         # Look for UDN embeds
1774         mobj = re.search(
1775             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage)
1776         if mobj is not None:
1777             return self.url_result(
1778                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1779
1780         # Look for Senate ISVP iframe
1781         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
1782         if senate_isvp_url:
1783             return self.url_result(senate_isvp_url, 'SenateISVP')
1784
1785         # Look for Dailymotion Cloud videos
1786         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
1787         if dmcloud_url:
1788             return self.url_result(dmcloud_url, 'DailymotionCloud')
1789
1790         # Look for OnionStudios embeds
1791         onionstudios_url = OnionStudiosIE._extract_url(webpage)
1792         if onionstudios_url:
1793             return self.url_result(onionstudios_url)
1794
1795         # Look for SnagFilms embeds
1796         snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage)
1797         if snagfilms_url:
1798             return self.url_result(snagfilms_url)
1799
1800         # Look for JWPlatform embeds
1801         jwplatform_url = JWPlatformIE._extract_url(webpage)
1802         if jwplatform_url:
1803             return self.url_result(jwplatform_url, 'JWPlatform')
1804
1805         # Look for ScreenwaveMedia embeds
1806         mobj = re.search(ScreenwaveMediaIE.EMBED_PATTERN, webpage)
1807         if mobj is not None:
1808             return self.url_result(unescapeHTML(mobj.group('url')), 'ScreenwaveMedia')
1809
1810         # Look for AdobeTVVideo embeds
1811         mobj = re.search(
1812             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
1813             webpage)
1814         if mobj is not None:
1815             return self.url_result(
1816                 self._proto_relative_url(unescapeHTML(mobj.group(1))),
1817                 'AdobeTVVideo')
1818
1819         def check_video(vurl):
1820             if YoutubeIE.suitable(vurl):
1821                 return True
1822             vpath = compat_urlparse.urlparse(vurl).path
1823             vext = determine_ext(vpath)
1824             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1825
1826         def filter_video(urls):
1827             return list(filter(check_video, urls))
1828
1829         # Start with something easy: JW Player in SWFObject
1830         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1831         if not found:
1832             # Look for gorilla-vid style embedding
1833             found = filter_video(re.findall(r'''(?sx)
1834                 (?:
1835                     jw_plugins|
1836                     JWPlayerOptions|
1837                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1838                 )
1839                 .*?
1840                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1841         if not found:
1842             # Broaden the search a little bit
1843             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1844         if not found:
1845             # Broaden the findall a little bit: JWPlayer JS loader
1846             found = filter_video(re.findall(
1847                 r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1848         if not found:
1849             # Flow player
1850             found = filter_video(re.findall(r'''(?xs)
1851                 flowplayer\("[^"]+",\s*
1852                     \{[^}]+?\}\s*,
1853                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1854                         ["']?url["']?\s*:\s*["']([^"']+)["']
1855             ''', webpage))
1856         if not found:
1857             # Cinerama player
1858             found = re.findall(
1859                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1860         if not found:
1861             # Try to find twitter cards info
1862             found = filter_video(re.findall(
1863                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1864         if not found:
1865             # We look for Open Graph info:
1866             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1867             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1868             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1869             if m_video_type is not None:
1870                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1871         if not found:
1872             # HTML5 video
1873             found = re.findall(r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1874         if not found:
1875             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1876             found = re.search(
1877                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1878                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1879                 webpage)
1880             if not found:
1881                 # Look also in Refresh HTTP header
1882                 refresh_header = head_response.headers.get('Refresh')
1883                 if refresh_header:
1884                     # In python 2 response HTTP headers are bytestrings
1885                     if sys.version_info < (3, 0) and isinstance(refresh_header, str):
1886                         refresh_header = refresh_header.decode('iso-8859-1')
1887                     found = re.search(REDIRECT_REGEX, refresh_header)
1888             if found:
1889                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
1890                 self.report_following_redirect(new_url)
1891                 return {
1892                     '_type': 'url',
1893                     'url': new_url,
1894                 }
1895         if not found:
1896             raise UnsupportedError(url)
1897
1898         entries = []
1899         for video_url in found:
1900             video_url = video_url.replace('\\/', '/')
1901             video_url = compat_urlparse.urljoin(url, video_url)
1902             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
1903
1904             # Sometimes, jwplayer extraction will result in a YouTube URL
1905             if YoutubeIE.suitable(video_url):
1906                 entries.append(self.url_result(video_url, 'Youtube'))
1907                 continue
1908
1909             # here's a fun little line of code for you:
1910             video_id = os.path.splitext(video_id)[0]
1911
1912             entry_info_dict = {
1913                 'id': video_id,
1914                 'uploader': video_uploader,
1915                 'title': video_title,
1916                 'age_limit': age_limit,
1917             }
1918
1919             ext = determine_ext(video_url)
1920             if ext == 'smil':
1921                 entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
1922             elif ext == 'xspf':
1923                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
1924             elif ext == 'm3u8':
1925                 entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
1926             else:
1927                 entry_info_dict['url'] = video_url
1928
1929             entries.append(entry_info_dict)
1930
1931         if len(entries) == 1:
1932             return entries[0]
1933         else:
1934             for num, e in enumerate(entries, start=1):
1935                 # 'url' results don't have a title
1936                 if e.get('title') is not None:
1937                     e['title'] = '%s (%d)' % (e['title'], num)
1938             return {
1939                 '_type': 'playlist',
1940                 'entries': entries,
1941             }