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

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