_ 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 .tnaflix import TNAFlixNetworkEmbedIE
  51 from .vimeo import VimeoIE
  52 from .dailymotion import DailymotionCloudIE
  53 from .onionstudios import OnionStudiosIE
  54 from .snagfilms import SnagFilmsEmbedIE
  55 from .screenwavemedia import ScreenwaveMediaIE
  56 from .mtv import MTVServicesEmbeddedIE
  57 from .pladform import PladformIE
  58 from .videomore import VideomoreIE
  59 from .googledrive import GoogleDriveIE
  60 from .jwplatform import JWPlatformIE
  61 from .digiteka import DigitekaIE
  62
  63
  64 class GenericIE(InfoExtractor):
  65     IE_DESC = 'Generic downloader that works on some sites'
  66     _VALID_URL = r'.*'
  67     IE_NAME = 'generic'
  68     _TESTS = [
  69         # Direct link to a video
  70         {
  71             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  72             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  73             'info_dict': {
  74                 'id': 'trailer',
  75                 'ext': 'mp4',
  76                 'title': 'trailer',
  77                 'upload_date': '20100513',
  78             }
  79         },
  80         # Direct link to media delivered compressed (until Accept-Encoding is *)
  81         {
  82             'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
  83             'md5': '128c42e68b13950268b648275386fc74',
  84             'info_dict': {
  85                 'id': 'FictionJunction-Parallel_Hearts',
  86                 'ext': 'flac',
  87                 'title': 'FictionJunction-Parallel_Hearts',
  88                 'upload_date': '20140522',
  89             },
  90             'expected_warnings': [
  91                 'URL could be a direct video link, returning it as such.'
  92             ]
  93         },
  94         # Direct download with broken HEAD
  95         {
  96             'url': 'http://ai-radio.org:8000/radio.opus',
  97             'info_dict': {
  98                 'id': 'radio',
  99                 'ext': 'opus',
 100                 'title': 'radio',
 101             },
 102             'params': {
 103                 'skip_download': True,  # infinite live stream
 104             },
 105             'expected_warnings': [
 106                 r'501.*Not Implemented'
 107             ],
 108         },
 109         # Direct link with incorrect MIME type
 110         {
 111             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 112             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 113             'info_dict': {
 114                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 115                 'id': '5_Lennart_Poettering_-_Systemd',
 116                 'ext': 'webm',
 117                 'title': '5_Lennart_Poettering_-_Systemd',
 118                 'upload_date': '20141120',
 119             },
 120             'expected_warnings': [
 121                 'URL could be a direct video link, returning it as such.'
 122             ]
 123         },
 124         # RSS feed
 125         {
 126             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 127             'info_dict': {
 128                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 129                 'title': 'Zero Punctuation',
 130                 'description': 're:.*groundbreaking video review series.*'
 131             },
 132             'playlist_mincount': 11,
 133         },
 134         # RSS feed with enclosure
 135         {
 136             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 137             'info_dict': {
 138                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 139                 'ext': 'm4v',
 140                 'upload_date': '20150228',
 141                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 142             }
 143         },
 144         # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
 145         {
 146             'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
 147             'info_dict': {
 148                 'id': 'smil',
 149                 'ext': 'mp4',
 150                 'title': 'Automatics, robotics and biocybernetics',
 151                 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
 152                 'upload_date': '20130627',
 153                 'formats': 'mincount:16',
 154                 'subtitles': 'mincount:1',
 155             },
 156             'params': {
 157                 'force_generic_extractor': True,
 158                 'skip_download': True,
 159             },
 160         },
 161         # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html
 162         {
 163             'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil',
 164             'info_dict': {
 165                 'id': 'hds',
 166                 'ext': 'flv',
 167                 'title': 'hds',
 168                 'formats': 'mincount:1',
 169             },
 170             'params': {
 171                 'skip_download': True,
 172             },
 173         },
 174         # SMIL from https://www.restudy.dk/video/play/id/1637
 175         {
 176             'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml',
 177             'info_dict': {
 178                 'id': 'video_1637',
 179                 'ext': 'flv',
 180                 'title': 'video_1637',
 181                 'formats': 'mincount:3',
 182             },
 183             'params': {
 184                 'skip_download': True,
 185             },
 186         },
 187         # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm
 188         {
 189             'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil',
 190             'info_dict': {
 191                 'id': 'smil-service',
 192                 'ext': 'flv',
 193                 'title': 'smil-service',
 194                 'formats': 'mincount:1',
 195             },
 196             'params': {
 197                 'skip_download': True,
 198             },
 199         },
 200         # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370
 201         {
 202             'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil',
 203             'info_dict': {
 204                 'id': '4719370',
 205                 'ext': 'mp4',
 206                 'title': '571de1fd-47bc-48db-abf9-238872a58d1f',
 207                 'formats': 'mincount:3',
 208             },
 209             'params': {
 210                 'skip_download': True,
 211             },
 212         },
 213         # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html
 214         {
 215             'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf',
 216             'info_dict': {
 217                 'id': 'mZlp2ctYIUEB',
 218                 'ext': 'mp4',
 219                 'title': 'Tikibad ontruimd wegens brand',
 220                 'description': 'md5:05ca046ff47b931f9b04855015e163a4',
 221                 'thumbnail': 're:^https?://.*\.jpg$',
 222                 'duration': 33,
 223             },
 224             'params': {
 225                 'skip_download': True,
 226             },
 227         },
 228         # MPD from http://dash-mse-test.appspot.com/media.html
 229         {
 230             'url': 'http://yt-dash-mse-test.commondatastorage.googleapis.com/media/car-20120827-manifest.mpd',
 231             'md5': '4b57baab2e30d6eb3a6a09f0ba57ef53',
 232             'info_dict': {
 233                 'id': 'car-20120827-manifest',
 234                 'ext': 'mp4',
 235                 'title': 'car-20120827-manifest',
 236                 'formats': 'mincount:9',
 237             },
 238             'params': {
 239                 'format': 'bestvideo',
 240             },
 241         },
 242         # google redirect
 243         {
 244             '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',
 245             'info_dict': {
 246                 'id': 'cmQHVoWB5FY',
 247                 'ext': 'mp4',
 248                 'upload_date': '20130224',
 249                 'uploader_id': 'TheVerge',
 250                 'description': 're:^Chris Ziegler takes a look at the\.*',
 251                 'uploader': 'The Verge',
 252                 'title': 'First Firefox OS phones side-by-side',
 253             },
 254             'params': {
 255                 'skip_download': False,
 256             }
 257         },
 258         {
 259             # redirect in Refresh HTTP header
 260             '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',
 261             'info_dict': {
 262                 'id': 'pO8h3EaFRdo',
 263                 'ext': 'mp4',
 264                 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
 265                 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5',
 266                 'upload_date': '20150917',
 267                 'uploader_id': 'brtvofficial',
 268                 'uploader': 'Boiler Room',
 269             },
 270             'params': {
 271                 'skip_download': False,
 272             },
 273         },
 274         {
 275             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
 276             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
 277             'info_dict': {
 278                 'id': '13601338388002',
 279                 'ext': 'mp4',
 280                 'uploader': 'www.hodiho.fr',
 281                 'title': 'R\u00e9gis plante sa Jeep',
 282             }
 283         },
 284         # bandcamp page with custom domain
 285         {
 286             'add_ie': ['Bandcamp'],
 287             'url': 'http://bronyrock.com/track/the-pony-mash',
 288             'info_dict': {
 289                 'id': '3235767654',
 290                 'ext': 'mp3',
 291                 'title': 'The Pony Mash',
 292                 'uploader': 'M_Pallante',
 293             },
 294             'skip': 'There is a limit of 200 free downloads / month for the test song',
 295         },
 296         # embedded brightcove video
 297         # it also tests brightcove videos that need to set the 'Referer' in the
 298         # http requests
 299         {
 300             'add_ie': ['BrightcoveLegacy'],
 301             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
 302             'info_dict': {
 303                 'id': '2765128793001',
 304                 'ext': 'mp4',
 305                 'title': 'Le cours de bourse : l’analyse technique',
 306                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
 307                 'uploader': 'BFM BUSINESS',
 308             },
 309             'params': {
 310                 'skip_download': True,
 311             },
 312         },
 313         {
 314             # https://github.com/rg3/youtube-dl/issues/2253
 315             'url': 'http://bcove.me/i6nfkrc3',
 316             'md5': '0ba9446db037002366bab3b3eb30c88c',
 317             'info_dict': {
 318                 'id': '3101154703001',
 319                 'ext': 'mp4',
 320                 'title': 'Still no power',
 321                 'uploader': 'thestar.com',
 322                 '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.',
 323             },
 324             'add_ie': ['BrightcoveLegacy'],
 325         },
 326         {
 327             'url': 'http://www.championat.com/video/football/v/87/87499.html',
 328             'md5': 'fb973ecf6e4a78a67453647444222983',
 329             'info_dict': {
 330                 'id': '3414141473001',
 331                 'ext': 'mp4',
 332                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 333                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 334                 'uploader': 'Championat',
 335             },
 336         },
 337         {
 338             # https://github.com/rg3/youtube-dl/issues/3541
 339             'add_ie': ['BrightcoveLegacy'],
 340             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 341             'info_dict': {
 342                 'id': '3866516442001',
 343                 'ext': 'mp4',
 344                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 345                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 346                 'uploader': 'SBS Broadcasting',
 347             },
 348             'skip': 'Restricted to Netherlands',
 349             'params': {
 350                 'skip_download': True,  # m3u8 download
 351             },
 352         },
 353         # ooyala video
 354         {
 355             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 356             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 357             'info_dict': {
 358                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 359                 'ext': 'mp4',
 360                 'title': '2cc213299525360.mov',  # that's what we get
 361                 'duration': 238.231,
 362             },
 363             'add_ie': ['Ooyala'],
 364         },
 365         {
 366             # ooyala video embedded with http://player.ooyala.com/iframe.js
 367             'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/',
 368             'info_dict': {
 369                 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB',
 370                 'ext': 'mp4',
 371                 'title': '"Steve Jobs: Man in the Machine" trailer',
 372                 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
 373                 'duration': 135.427,
 374             },
 375             'params': {
 376                 'skip_download': True,
 377             },
 378         },
 379         # multiple ooyala embeds on SBN network websites
 380         {
 381             'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 382             'info_dict': {
 383                 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 384                 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
 385             },
 386             'playlist_mincount': 3,
 387             'params': {
 388                 'skip_download': True,
 389             },
 390             'add_ie': ['Ooyala'],
 391         },
 392         # embed.ly video
 393         {
 394             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 395             'info_dict': {
 396                 'id': '9ODmcdjQcHQ',
 397                 'ext': 'mp4',
 398                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 399                 'upload_date': '20140225',
 400                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 401                 'uploader': 'Tested',
 402                 'uploader_id': 'testedcom',
 403             },
 404             # No need to test YoutubeIE here
 405             'params': {
 406                 'skip_download': True,
 407             },
 408         },
 409         # funnyordie embed
 410         {
 411             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 412             'info_dict': {
 413                 'id': '18e820ec3f',
 414                 'ext': 'mp4',
 415                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 416                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 417             },
 418         },
 419         # RUTV embed
 420         {
 421             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 422             'info_dict': {
 423                 'id': '776940',
 424                 'ext': 'mp4',
 425                 'title': 'Охотское море стало целиком российским',
 426                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 427             },
 428             'params': {
 429                 # m3u8 download
 430                 'skip_download': True,
 431             },
 432         },
 433         # TVC embed
 434         {
 435             '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/',
 436             'info_dict': {
 437                 'id': '55304',
 438                 'ext': 'mp4',
 439                 'title': 'Дошкольное воспитание',
 440             },
 441         },
 442         # SportBox embed
 443         {
 444             'url': 'http://www.vestifinance.ru/articles/25753',
 445             'info_dict': {
 446                 'id': '25753',
 447                 'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
 448             },
 449             'playlist': [{
 450                 'info_dict': {
 451                     'id': '370908',
 452                     'title': 'Госзаказ. День 3',
 453                     'ext': 'mp4',
 454                 }
 455             }, {
 456                 'info_dict': {
 457                     'id': '370905',
 458                     'title': 'Госзаказ. День 2',
 459                     'ext': 'mp4',
 460                 }
 461             }, {
 462                 'info_dict': {
 463                     'id': '370902',
 464                     'title': 'Госзаказ. День 1',
 465                     'ext': 'mp4',
 466                 }
 467             }],
 468             'params': {
 469                 # m3u8 download
 470                 'skip_download': True,
 471             },
 472         },
 473         # Myvi.ru embed
 474         {
 475             'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
 476             'info_dict': {
 477                 'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
 478                 'ext': 'mp4',
 479                 'title': 'Ужастики, русский трейлер (2015)',
 480                 'thumbnail': 're:^https?://.*\.jpg$',
 481                 'duration': 153,
 482             }
 483         },
 484         # XHamster embed
 485         {
 486             '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',
 487             'info_dict': {
 488                 'id': 'showthread',
 489                 'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
 490             },
 491             'playlist_mincount': 7,
 492         },
 493         # Embedded TED video
 494         {
 495             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 496             'md5': '65fdff94098e4a607385a60c5177c638',
 497             'info_dict': {
 498                 'id': '1969',
 499                 'ext': 'mp4',
 500                 'title': 'Hidden miracles of the natural world',
 501                 'uploader': 'Louie Schwartzberg',
 502                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 503             }
 504         },
 505         # Embedded Ustream video
 506         {
 507             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 508             'md5': '27b99cdb639c9b12a79bca876a073417',
 509             'info_dict': {
 510                 'id': '45734260',
 511                 'ext': 'flv',
 512                 'uploader': 'AU SPA:  The NSA and Privacy',
 513                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 514             }
 515         },
 516         # nowvideo embed hidden behind percent encoding
 517         {
 518             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 519             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 520             'info_dict': {
 521                 'id': '06e53103ca9aa',
 522                 'ext': 'flv',
 523                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 524                 'description': 'No description',
 525             },
 526         },
 527         # arte embed
 528         {
 529             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 530             'md5': '7653032cbb25bf6c80d80f217055fa43',
 531             'info_dict': {
 532                 'id': '048195-004_PLUS7-F',
 533                 'ext': 'flv',
 534                 'title': 'X:enius',
 535                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 536                 'upload_date': '20140320',
 537             },
 538             'params': {
 539                 'skip_download': 'Requires rtmpdump'
 540             }
 541         },
 542         # francetv embed
 543         {
 544             'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
 545             'info_dict': {
 546                 'id': 'EV_30231',
 547                 'ext': 'mp4',
 548                 'title': 'Alcaline, le concert avec Calogero',
 549                 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
 550                 'upload_date': '20150226',
 551                 'timestamp': 1424989860,
 552                 'duration': 5400,
 553             },
 554             'params': {
 555                 # m3u8 downloads
 556                 'skip_download': True,
 557             },
 558             'expected_warnings': [
 559                 'Forbidden'
 560             ]
 561         },
 562         # Condé Nast embed
 563         {
 564             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 565             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 566             'info_dict': {
 567                 'id': '53501be369702d3275860000',
 568                 'ext': 'mp4',
 569                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 570             }
 571         },
 572         # Dailymotion embed
 573         {
 574             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 575             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 576             'info_dict': {
 577                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 578                 'ext': 'mp4',
 579                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 580                 'uploader': 'Spi0n',
 581             },
 582             'add_ie': ['Dailymotion'],
 583         },
 584         # YouTube embed
 585         {
 586             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 587             'info_dict': {
 588                 'id': 'FXRb4ykk4S0',
 589                 'ext': 'mp4',
 590                 'title': 'The NBL Auction 2014',
 591                 'uploader': 'BADMINTON England',
 592                 'uploader_id': 'BADMINTONEvents',
 593                 'upload_date': '20140603',
 594                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 595             },
 596             'add_ie': ['Youtube'],
 597             'params': {
 598                 'skip_download': True,
 599             }
 600         },
 601         # MTVSercices embed
 602         {
 603             'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
 604             'md5': '35727f82f58c76d996fc188f9755b0d5',
 605             'info_dict': {
 606                 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
 607                 'ext': 'mp4',
 608                 'title': 'Review',
 609                 'description': 'Mario\'s life in the fast lane has never looked so good.',
 610             },
 611         },
 612         # YouTube embed via <data-embed-url="">
 613         {
 614             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 615             'info_dict': {
 616                 'id': '4vAffPZIT44',
 617                 'ext': 'mp4',
 618                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 619                 'uploader': 'Gameloft',
 620                 'uploader_id': 'gameloft',
 621                 'upload_date': '20140828',
 622                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 623             },
 624             'params': {
 625                 'skip_download': True,
 626             }
 627         },
 628         # Camtasia studio
 629         {
 630             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 631             'playlist': [{
 632                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 633                 'info_dict': {
 634                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 635                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 636                     'ext': 'flv',
 637                     'duration': 2235.90,
 638                 }
 639             }, {
 640                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 641                 'info_dict': {
 642                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 643                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 644                     'ext': 'flv',
 645                     'duration': 2235.93,
 646                 }
 647             }],
 648             'info_dict': {
 649                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 650             }
 651         },
 652         # Flowplayer
 653         {
 654             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 655             'md5': '9d65602bf31c6e20014319c7d07fba27',
 656             'info_dict': {
 657                 'id': '5123ea6d5e5a7',
 658                 'ext': 'mp4',
 659                 'age_limit': 18,
 660                 'uploader': 'www.handjobhub.com',
 661                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 662             }
 663         },
 664         # Multiple brightcove videos
 665         # https://github.com/rg3/youtube-dl/issues/2283
 666         {
 667             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 668             'info_dict': {
 669                 'id': 'always-never',
 670                 'title': 'Always / Never - The New Yorker',
 671             },
 672             'playlist_count': 3,
 673             'params': {
 674                 'extract_flat': False,
 675                 'skip_download': True,
 676             }
 677         },
 678         # MLB embed
 679         {
 680             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 681             'md5': '96f09a37e44da40dd083e12d9a683327',
 682             'info_dict': {
 683                 'id': '33322633',
 684                 'ext': 'mp4',
 685                 'title': 'Ump changes call to ball',
 686                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 687                 'duration': 48,
 688                 'timestamp': 1401537900,
 689                 'upload_date': '20140531',
 690                 'thumbnail': 're:^https?://.*\.jpg$',
 691             },
 692         },
 693         # Wistia embed
 694         {
 695             'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 696             'md5': '8788b683c777a5cf25621eaf286d0c23',
 697             'info_dict': {
 698                 'id': '1cfaf6b7ea',
 699                 'ext': 'mov',
 700                 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
 701                 'duration': 643.0,
 702                 'filesize': 182808282,
 703                 'uploader': 'education-portal.com',
 704             },
 705         },
 706         {
 707             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 708             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 709             'info_dict': {
 710                 'id': 'uxjb0lwrcz',
 711                 'ext': 'mp4',
 712                 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
 713                 'duration': 1715.0,
 714                 'uploader': 'thoughtworks.wistia.com',
 715             },
 716         },
 717         # Soundcloud embed
 718         {
 719             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 720             'info_dict': {
 721                 'id': '174391317',
 722                 'ext': 'mp3',
 723                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 724                 'uploader': 'Sophos Security',
 725                 'title': 'Chet Chat 171 - Oct 29, 2014',
 726                 'upload_date': '20141029',
 727             }
 728         },
 729         # Livestream embed
 730         {
 731             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 732             'info_dict': {
 733                 'id': '67864563',
 734                 'ext': 'flv',
 735                 'upload_date': '20141112',
 736                 'title': 'Rosetta #CometLanding webcast HL 10',
 737             }
 738         },
 739         # LazyYT
 740         {
 741             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
 742             'info_dict': {
 743                 'id': '1986',
 744                 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
 745             },
 746             'playlist_mincount': 2,
 747         },
 748         # Cinchcast embed
 749         {
 750             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
 751             'info_dict': {
 752                 'id': '7141703',
 753                 'ext': 'mp3',
 754                 'upload_date': '20141126',
 755                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
 756             }
 757         },
 758         # Cinerama player
 759         {
 760             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
 761             'info_dict': {
 762                 'id': '730m_DandD_1901_512k',
 763                 'ext': 'mp4',
 764                 'uploader': 'www.abc.net.au',
 765                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
 766             }
 767         },
 768         # embedded viddler video
 769         {
 770             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
 771             'info_dict': {
 772                 'id': '4d03aad9',
 773                 'ext': 'mp4',
 774                 'uploader': 'deadspin',
 775                 'title': 'WALL-TO-GORTAT',
 776                 'timestamp': 1422285291,
 777                 'upload_date': '20150126',
 778             },
 779             'add_ie': ['Viddler'],
 780         },
 781         # Libsyn embed
 782         {
 783             'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
 784             'info_dict': {
 785                 'id': '3377616',
 786                 'ext': 'mp3',
 787                 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
 788                 'description': 'md5:601cb790edd05908957dae8aaa866465',
 789                 'upload_date': '20150220',
 790             },
 791         },
 792         # jwplayer YouTube
 793         {
 794             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
 795             'info_dict': {
 796                 'id': 'Mrj4DVp2zeA',
 797                 'ext': 'mp4',
 798                 'upload_date': '20150212',
 799                 'uploader': 'The National Archives UK',
 800                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
 801                 'uploader_id': 'NationalArchives08',
 802                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
 803             },
 804         },
 805         # rtl.nl embed
 806         {
 807             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
 808             'playlist_mincount': 5,
 809             'info_dict': {
 810                 'id': 'aanslagen-kopenhagen',
 811                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
 812             }
 813         },
 814         # Zapiks embed
 815         {
 816             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
 817             'info_dict': {
 818                 'id': '118046',
 819                 'ext': 'mp4',
 820                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
 821             }
 822         },
 823         # Kaltura embed
 824         {
 825             'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
 826             'info_dict': {
 827                 'id': '1_eergr3h1',
 828                 'ext': 'mp4',
 829                 'upload_date': '20150226',
 830                 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
 831                 'timestamp': int,
 832                 'title': 'John Carlson Postgame 2/25/15',
 833             },
 834         },
 835         # Kaltura embed (different embed code)
 836         {
 837             'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
 838             'info_dict': {
 839                 'id': '1_a52wc67y',
 840                 'ext': 'flv',
 841                 'upload_date': '20150127',
 842                 'uploader_id': 'PremierMedia',
 843                 'timestamp': int,
 844                 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
 845             },
 846         },
 847         # Kaltura embed protected with referrer
 848         {
 849             'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero',
 850             'info_dict': {
 851                 'id': '1_g4fbemnq',
 852                 'ext': 'mp4',
 853                 'title': 'Violetta - Achter De Schermen - Ruggero',
 854                 'description': 'Achter de schermen met Ruggero',
 855                 'timestamp': 1435133761,
 856                 'upload_date': '20150624',
 857                 'uploader_id': 'echojecka',
 858             },
 859         },
 860         # Eagle.Platform embed (generic URL)
 861         {
 862             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
 863             'info_dict': {
 864                 'id': '227304',
 865                 'ext': 'mp4',
 866                 'title': 'Навальный вышел на свободу',
 867                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
 868                 'thumbnail': 're:^https?://.*\.jpg$',
 869                 'duration': 87,
 870                 'view_count': int,
 871                 'age_limit': 0,
 872             },
 873         },
 874         # ClipYou (Eagle.Platform) embed (custom URL)
 875         {
 876             'url': 'http://muz-tv.ru/play/7129/',
 877             'info_dict': {
 878                 'id': '12820',
 879                 'ext': 'mp4',
 880                 'title': "'O Sole Mio",
 881                 'thumbnail': 're:^https?://.*\.jpg$',
 882                 'duration': 216,
 883                 'view_count': int,
 884             },
 885         },
 886         # Pladform embed
 887         {
 888             'url': 'http://muz-tv.ru/kinozal/view/7400/',
 889             'info_dict': {
 890                 'id': '100183293',
 891                 'ext': 'mp4',
 892                 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
 893                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
 894                 'thumbnail': 're:^https?://.*\.jpg$',
 895                 'duration': 694,
 896                 'age_limit': 0,
 897             },
 898         },
 899         # Playwire embed
 900         {
 901             'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
 902             'info_dict': {
 903                 'id': '3519514',
 904                 'ext': 'mp4',
 905                 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
 906                 'thumbnail': 're:^https?://.*\.png$',
 907                 'duration': 45.115,
 908             },
 909         },
 910         # 5min embed
 911         {
 912             'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
 913             'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
 914             'info_dict': {
 915                 'id': '518726732',
 916                 'ext': 'mp4',
 917                 'title': 'Facebook Creates "On This Day" | Crunch Report',
 918             },
 919         },
 920         # SVT embed
 921         {
 922             'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 923             'info_dict': {
 924                 'id': '2900353',
 925                 'ext': 'flv',
 926                 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
 927                 'duration': 27,
 928                 'age_limit': 0,
 929             },
 930         },
 931         # Crooks and Liars embed
 932         {
 933             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
 934             'info_dict': {
 935                 'id': '8RUoRhRi',
 936                 'ext': 'mp4',
 937                 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
 938                 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
 939                 'timestamp': 1428207000,
 940                 'upload_date': '20150405',
 941                 'uploader': 'Heather',
 942             },
 943         },
 944         # Crooks and Liars external embed
 945         {
 946             'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
 947             'info_dict': {
 948                 'id': 'MTE3MjUtMzQ2MzA',
 949                 'ext': 'mp4',
 950                 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
 951                 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
 952                 'timestamp': 1265032391,
 953                 'upload_date': '20100201',
 954                 'uploader': 'Heather',
 955             },
 956         },
 957         # NBC Sports vplayer embed
 958         {
 959             'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
 960             'info_dict': {
 961                 'id': 'ln7x1qSThw4k',
 962                 'ext': 'flv',
 963                 'title': "PFT Live: New leader in the 'new-look' defense",
 964                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
 965             },
 966         },
 967         # UDN embed
 968         {
 969             'url': 'http://www.udn.com/news/story/7314/822787',
 970             'md5': 'fd2060e988c326991037b9aff9df21a6',
 971             'info_dict': {
 972                 'id': '300346',
 973                 'ext': 'mp4',
 974                 'title': '中一中男師變性 全校師生力挺',
 975                 'thumbnail': 're:^https?://.*\.jpg$',
 976             }
 977         },
 978         # Ooyala embed
 979         {
 980             'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
 981             'info_dict': {
 982                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
 983                 'ext': 'mp4',
 984                 'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.',
 985                 'title': 'This is what separates the Excel masters from the wannabes',
 986                 'duration': 191.933,
 987             },
 988             'params': {
 989                 # m3u8 downloads
 990                 'skip_download': True,
 991             }
 992         },
 993         # Contains a SMIL manifest
 994         {
 995             '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',
 996             'info_dict': {
 997                 'id': 'file',
 998                 'ext': 'flv',
 999                 'title': '+ Football: Lottery Champions League Europe',
1000                 'uploader': 'www.telewebion.com',
1001             },
1002             'params': {
1003                 # rtmpe downloads
1004                 'skip_download': True,
1005             }
1006         },
1007         # Brightcove URL in single quotes
1008         {
1009             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
1010             'md5': '4ae374f1f8b91c889c4b9203c8c752af',
1011             'info_dict': {
1012                 'id': '4255764656001',
1013                 'ext': 'mp4',
1014                 'title': 'SN Presents: Russell Martin, World Citizen',
1015                 '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.',
1016                 'uploader': 'Rogers Sportsnet',
1017             },
1018         },
1019         # Dailymotion Cloud video
1020         {
1021             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
1022             'md5': '49444254273501a64675a7e68c502681',
1023             'info_dict': {
1024                 'id': '5585de919473990de4bee11b',
1025                 'ext': 'mp4',
1026                 'title': 'Le débat',
1027                 'thumbnail': 're:^https?://.*\.jpe?g$',
1028             }
1029         },
1030         # OnionStudios embed
1031         {
1032             'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
1033             'info_dict': {
1034                 'id': '2855',
1035                 'ext': 'mp4',
1036                 'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
1037                 'thumbnail': 're:^https?://.*\.jpe?g$',
1038                 'uploader': 'ClickHole',
1039                 'uploader_id': 'clickhole',
1040             }
1041         },
1042         # SnagFilms embed
1043         {
1044             'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
1045             'info_dict': {
1046                 'id': '74849a00-85a9-11e1-9660-123139220831',
1047                 'ext': 'mp4',
1048                 'title': '#whilewewatch',
1049             }
1050         },
1051         # AdobeTVVideo embed
1052         {
1053             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
1054             'md5': '43662b577c018ad707a63766462b1e87',
1055             'info_dict': {
1056                 'id': '2456',
1057                 'ext': 'mp4',
1058                 'title': 'New experience with Acrobat DC',
1059                 'description': 'New experience with Acrobat DC',
1060                 'duration': 248.667,
1061             },
1062         },
1063         # ScreenwaveMedia embed
1064         {
1065             'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1',
1066             'md5': '24ace5baba0d35d55c6810b51f34e9e0',
1067             'info_dict': {
1068                 'id': 'cinemasnob-55d26273809dd',
1069                 'ext': 'mp4',
1070                 'title': 'cinemasnob',
1071             },
1072         },
1073         # BrightcoveInPageEmbed embed
1074         {
1075             'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
1076             'info_dict': {
1077                 'id': '4238694884001',
1078                 'ext': 'flv',
1079                 'title': 'Tabletop: Dread, Last Thoughts',
1080                 'description': 'Tabletop: Dread, Last Thoughts',
1081                 'duration': 51690,
1082             },
1083         },
1084         # JWPlayer with M3U8
1085         {
1086             'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video',
1087             'info_dict': {
1088                 'id': 'playlist',
1089                 'ext': 'mp4',
1090                 'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ',
1091                 'uploader': 'ren.tv',
1092             },
1093             'params': {
1094                 # m3u8 downloads
1095                 'skip_download': True,
1096             }
1097         }
1098     ]
1099
1100     def report_following_redirect(self, new_url):
1101         """Report information extraction."""
1102         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
1103
1104     def _extract_rss(self, url, video_id, doc):
1105         playlist_title = doc.find('./channel/title').text
1106         playlist_desc_el = doc.find('./channel/description')
1107         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
1108
1109         entries = []
1110         for it in doc.findall('./channel/item'):
1111             next_url = xpath_text(it, 'link', fatal=False)
1112             if not next_url:
1113                 enclosure_nodes = it.findall('./enclosure')
1114                 for e in enclosure_nodes:
1115                     next_url = e.attrib.get('url')
1116                     if next_url:
1117                         break
1118
1119             if not next_url:
1120                 continue
1121
1122             entries.append({
1123                 '_type': 'url',
1124                 'url': next_url,
1125                 'title': it.find('title').text,
1126             })
1127
1128         return {
1129             '_type': 'playlist',
1130             'id': url,
1131             'title': playlist_title,
1132             'description': playlist_desc,
1133             'entries': entries,
1134         }
1135
1136     def _extract_camtasia(self, url, video_id, webpage):
1137         """ Returns None if no camtasia video can be found. """
1138
1139         camtasia_cfg = self._search_regex(
1140             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
1141             webpage, 'camtasia configuration file', default=None)
1142         if camtasia_cfg is None:
1143             return None
1144
1145         title = self._html_search_meta('DC.title', webpage, fatal=True)
1146
1147         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
1148         camtasia_cfg = self._download_xml(
1149             camtasia_url, video_id,
1150             note='Downloading camtasia configuration',
1151             errnote='Failed to download camtasia configuration')
1152         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
1153
1154         entries = []
1155         for n in fileset_node.getchildren():
1156             url_n = n.find('./uri')
1157             if url_n is None:
1158                 continue
1159
1160             entries.append({
1161                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
1162                 'title': '%s - %s' % (title, n.tag),
1163                 'url': compat_urlparse.urljoin(url, url_n.text),
1164                 'duration': float_or_none(n.find('./duration').text),
1165             })
1166
1167         return {
1168             '_type': 'playlist',
1169             'entries': entries,
1170             'title': title,
1171         }
1172
1173     def _real_extract(self, url):
1174         if url.startswith('//'):
1175             return {
1176                 '_type': 'url',
1177                 'url': self.http_scheme() + url,
1178             }
1179
1180         parsed_url = compat_urlparse.urlparse(url)
1181         if not parsed_url.scheme:
1182             default_search = self._downloader.params.get('default_search')
1183             if default_search is None:
1184                 default_search = 'fixup_error'
1185
1186             if default_search in ('auto', 'auto_warning', 'fixup_error'):
1187                 if '/' in url:
1188                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
1189                     return self.url_result('http://' + url)
1190                 elif default_search != 'fixup_error':
1191                     if default_search == 'auto_warning':
1192                         if re.match(r'^(?:url|URL)$', url):
1193                             raise ExtractorError(
1194                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
1195                                 expected=True)
1196                         else:
1197                             self._downloader.report_warning(
1198                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
1199                     return self.url_result('ytsearch:' + url)
1200
1201             if default_search in ('error', 'fixup_error'):
1202                 raise ExtractorError(
1203                     '%r is not a valid URL. '
1204                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
1205                     % (url, url), expected=True)
1206             else:
1207                 if ':' not in default_search:
1208                     default_search += ':'
1209                 return self.url_result(default_search + url)
1210
1211         url, smuggled_data = unsmuggle_url(url)
1212         force_videoid = None
1213         is_intentional = smuggled_data and smuggled_data.get('to_generic')
1214         if smuggled_data and 'force_videoid' in smuggled_data:
1215             force_videoid = smuggled_data['force_videoid']
1216             video_id = force_videoid
1217         else:
1218             video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
1219
1220         self.to_screen('%s: Requesting header' % video_id)
1221
1222         head_req = HEADRequest(url)
1223         head_response = self._request_webpage(
1224             head_req, video_id,
1225             note=False, errnote='Could not send HEAD request to %s' % url,
1226             fatal=False)
1227
1228         if head_response is not False:
1229             # Check for redirect
1230             new_url = head_response.geturl()
1231             if url != new_url:
1232                 self.report_following_redirect(new_url)
1233                 if force_videoid:
1234                     new_url = smuggle_url(
1235                         new_url, {'force_videoid': force_videoid})
1236                 return self.url_result(new_url)
1237
1238         full_response = None
1239         if head_response is False:
1240             request = sanitized_Request(url)
1241             request.add_header('Accept-Encoding', '*')
1242             full_response = self._request_webpage(request, video_id)
1243             head_response = full_response
1244
1245         # Check for direct link to a video
1246         content_type = head_response.headers.get('Content-Type', '')
1247         m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>.+)$', content_type)
1248         if m:
1249             upload_date = unified_strdate(
1250                 head_response.headers.get('Last-Modified'))
1251             formats = []
1252             if m.group('format_id').endswith('mpegurl'):
1253                 formats = self._extract_m3u8_formats(url, video_id, 'mp4')
1254             else:
1255                 formats = [{
1256                     'format_id': m.group('format_id'),
1257                     'url': url,
1258                     'vcodec': 'none' if m.group('type') == 'audio' else None
1259                 }]
1260             return {
1261                 'id': video_id,
1262                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1263                 'direct': True,
1264                 'formats': formats,
1265                 'upload_date': upload_date,
1266             }
1267
1268         if not self._downloader.params.get('test', False) and not is_intentional:
1269             force = self._downloader.params.get('force_generic_extractor', False)
1270             self._downloader.report_warning(
1271                 '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
1272
1273         if not full_response:
1274             request = sanitized_Request(url)
1275             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
1276             # making it impossible to download only chunk of the file (yet we need only 512kB to
1277             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
1278             # that will always result in downloading the whole file that is not desirable.
1279             # Therefore for extraction pass we have to override Accept-Encoding to any in order
1280             # to accept raw bytes and being able to download only a chunk.
1281             # It may probably better to solve this by checking Content-Type for application/octet-stream
1282             # after HEAD request finishes, but not sure if we can rely on this.
1283             request.add_header('Accept-Encoding', '*')
1284             full_response = self._request_webpage(request, video_id)
1285
1286         # Maybe it's a direct link to a video?
1287         # Be careful not to download the whole thing!
1288         first_bytes = full_response.read(512)
1289         if not is_html(first_bytes):
1290             self._downloader.report_warning(
1291                 'URL could be a direct video link, returning it as such.')
1292             upload_date = unified_strdate(
1293                 head_response.headers.get('Last-Modified'))
1294             return {
1295                 'id': video_id,
1296                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1297                 'direct': True,
1298                 'url': url,
1299                 'upload_date': upload_date,
1300             }
1301
1302         webpage = self._webpage_read_content(
1303             full_response, url, video_id, prefix=first_bytes)
1304
1305         self.report_extraction(video_id)
1306
1307         # Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest?
1308         try:
1309             doc = compat_etree_fromstring(webpage.encode('utf-8'))
1310             if doc.tag == 'rss':
1311                 return self._extract_rss(url, video_id, doc)
1312             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
1313                 return self._parse_smil(doc, url, video_id)
1314             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
1315                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
1316             elif re.match(r'(?i)^(?:{[^}]+})?MPD$', doc.tag):
1317                 return {
1318                     'id': video_id,
1319                     'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1320                     'formats': self._parse_mpd_formats(
1321                         doc, video_id, mpd_base_url=url.rpartition('/')[0]),
1322                 }
1323         except compat_xml_parse_error:
1324             pass
1325
1326         # Is it a Camtasia project?
1327         camtasia_res = self._extract_camtasia(url, video_id, webpage)
1328         if camtasia_res is not None:
1329             return camtasia_res
1330
1331         # Sometimes embedded video player is hidden behind percent encoding
1332         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1333         # Unescaping the whole page allows to handle those cases in a generic way
1334         webpage = compat_urllib_parse_unquote(webpage)
1335
1336         # it's tempting to parse this further, but you would
1337         # have to take into account all the variations like
1338         #   Video Title - Site Name
1339         #   Site Name | Video Title
1340         #   Video Title - Tagline | Site Name
1341         # and so on and so forth; it's just not practical
1342         video_title = self._html_search_regex(
1343             r'(?s)<title>(.*?)</title>', webpage, 'video title',
1344             default='video')
1345
1346         # Try to detect age limit automatically
1347         age_limit = self._rta_search(webpage)
1348         # And then there are the jokers who advertise that they use RTA,
1349         # but actually don't.
1350         AGE_LIMIT_MARKERS = [
1351             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1352         ]
1353         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1354             age_limit = 18
1355
1356         # video uploader is domain name
1357         video_uploader = self._search_regex(
1358             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1359
1360         # Helper method
1361         def _playlist_from_matches(matches, getter=None, ie=None):
1362             urlrs = orderedSet(
1363                 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1364                 for m in matches)
1365             return self.playlist_result(
1366                 urlrs, playlist_id=video_id, playlist_title=video_title)
1367
1368         # Look for Brightcove Legacy Studio embeds
1369         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
1370         if bc_urls:
1371             self.to_screen('Brightcove video detected.')
1372             entries = [{
1373                 '_type': 'url',
1374                 'url': smuggle_url(bc_url, {'Referer': url}),
1375                 'ie_key': 'BrightcoveLegacy'
1376             } for bc_url in bc_urls]
1377
1378             return {
1379                 '_type': 'playlist',
1380                 'title': video_title,
1381                 'id': video_id,
1382                 'entries': entries,
1383             }
1384
1385         # Look for Brightcove New Studio embeds
1386         bc_urls = BrightcoveNewIE._extract_urls(webpage)
1387         if bc_urls:
1388             return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
1389
1390         # Look for embedded rtl.nl player
1391         matches = re.findall(
1392             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1393             webpage)
1394         if matches:
1395             return _playlist_from_matches(matches, ie='RtlNl')
1396
1397         vimeo_url = VimeoIE._extract_vimeo_url(url, webpage)
1398         if vimeo_url is not None:
1399             return self.url_result(vimeo_url)
1400
1401         vid_me_embed_url = self._search_regex(
1402             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
1403             webpage, 'vid.me embed', default=None)
1404         if vid_me_embed_url is not None:
1405             return self.url_result(vid_me_embed_url, 'Vidme')
1406
1407         # Look for embedded YouTube player
1408         matches = re.findall(r'''(?x)
1409             (?:
1410                 <iframe[^>]+?src=|
1411                 data-video-url=|
1412                 <embed[^>]+?src=|
1413                 embedSWF\(?:\s*|
1414                 new\s+SWFObject\(
1415             )
1416             (["\'])
1417                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1418                 (?:embed|v|p)/.+?)
1419             \1''', webpage)
1420         if matches:
1421             return _playlist_from_matches(
1422                 matches, lambda m: unescapeHTML(m[1]))
1423
1424         # Look for lazyYT YouTube embed
1425         matches = re.findall(
1426             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1427         if matches:
1428             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1429
1430         # Look for embedded Dailymotion player
1431         matches = re.findall(
1432             r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
1433         if matches:
1434             return _playlist_from_matches(
1435                 matches, lambda m: unescapeHTML(m[1]))
1436
1437         # Look for embedded Dailymotion playlist player (#3822)
1438         m = re.search(
1439             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1440         if m:
1441             playlists = re.findall(
1442                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1443             if playlists:
1444                 return _playlist_from_matches(
1445                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1446
1447         # Look for embedded Wistia player
1448         match = re.search(
1449             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1450         if match:
1451             embed_url = self._proto_relative_url(
1452                 unescapeHTML(match.group('url')))
1453             return {
1454                 '_type': 'url_transparent',
1455                 'url': embed_url,
1456                 'ie_key': 'Wistia',
1457                 'uploader': video_uploader,
1458                 'title': video_title,
1459                 'id': video_id,
1460             }
1461
1462         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1463         if match:
1464             return {
1465                 '_type': 'url_transparent',
1466                 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1467                 'ie_key': 'Wistia',
1468                 'uploader': video_uploader,
1469                 'title': video_title,
1470                 'id': match.group('id')
1471             }
1472
1473         # Look for SVT player
1474         svt_url = SVTIE._extract_url(webpage)
1475         if svt_url:
1476             return self.url_result(svt_url, 'SVT')
1477
1478         # Look for embedded condenast player
1479         matches = re.findall(
1480             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1481             webpage)
1482         if matches:
1483             return {
1484                 '_type': 'playlist',
1485                 'entries': [{
1486                     '_type': 'url',
1487                     'ie_key': 'CondeNast',
1488                     'url': ma,
1489                 } for ma in matches],
1490                 'title': video_title,
1491                 'id': video_id,
1492             }
1493
1494         # Look for Bandcamp pages with custom domain
1495         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1496         if mobj is not None:
1497             burl = unescapeHTML(mobj.group(1))
1498             # Don't set the extractor because it can be a track url or an album
1499             return self.url_result(burl)
1500
1501         # Look for embedded Vevo player
1502         mobj = re.search(
1503             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1504         if mobj is not None:
1505             return self.url_result(mobj.group('url'))
1506
1507         # Look for embedded Viddler player
1508         mobj = re.search(
1509             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1510             webpage)
1511         if mobj is not None:
1512             return self.url_result(mobj.group('url'))
1513
1514         # Look for NYTimes player
1515         mobj = re.search(
1516             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1517             webpage)
1518         if mobj is not None:
1519             return self.url_result(mobj.group('url'))
1520
1521         # Look for Libsyn player
1522         mobj = re.search(
1523             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1524         if mobj is not None:
1525             return self.url_result(mobj.group('url'))
1526
1527         # Look for Ooyala videos
1528         mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1529                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1530                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1531                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1532         if mobj is not None:
1533             return OoyalaIE._build_url_result(smuggle_url(mobj.group('ec'), {'domain': url}))
1534
1535         # Look for multiple Ooyala embeds on SBN network websites
1536         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1537         if mobj is not None:
1538             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1539             if embeds:
1540                 return _playlist_from_matches(
1541                     embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
1542
1543         # Look for Aparat videos
1544         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1545         if mobj is not None:
1546             return self.url_result(mobj.group(1), 'Aparat')
1547
1548         # Look for MPORA videos
1549         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1550         if mobj is not None:
1551             return self.url_result(mobj.group(1), 'Mpora')
1552
1553         # Look for embedded NovaMov-based player
1554         mobj = re.search(
1555             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1556                     (?P<url>http://(?:(?:embed|www)\.)?
1557                         (?:novamov\.com|
1558                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
1559                            videoweed\.(?:es|com)|
1560                            movshare\.(?:net|sx|ag)|
1561                            divxstage\.(?:eu|net|ch|co|at|ag))
1562                         /embed\.php.+?)\1''', webpage)
1563         if mobj is not None:
1564             return self.url_result(mobj.group('url'))
1565
1566         # Look for embedded Facebook player
1567         mobj = re.search(
1568             r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1569         if mobj is not None:
1570             return self.url_result(mobj.group('url'), 'Facebook')
1571
1572         # Look for embedded VK player
1573         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1574         if mobj is not None:
1575             return self.url_result(mobj.group('url'), 'VK')
1576
1577         # Look for embedded Odnoklassniki player
1578         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1', webpage)
1579         if mobj is not None:
1580             return self.url_result(mobj.group('url'), 'Odnoklassniki')
1581
1582         # Look for embedded ivi player
1583         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1584         if mobj is not None:
1585             return self.url_result(mobj.group('url'), 'Ivi')
1586
1587         # Look for embedded Huffington Post player
1588         mobj = re.search(
1589             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1590         if mobj is not None:
1591             return self.url_result(mobj.group('url'), 'HuffPost')
1592
1593         # Look for embed.ly
1594         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1595         if mobj is not None:
1596             return self.url_result(mobj.group('url'))
1597         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1598         if mobj is not None:
1599             return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
1600
1601         # Look for funnyordie embed
1602         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1603         if matches:
1604             return _playlist_from_matches(
1605                 matches, getter=unescapeHTML, ie='FunnyOrDie')
1606
1607         # Look for BBC iPlayer embed
1608         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1609         if matches:
1610             return _playlist_from_matches(matches, ie='BBCCoUk')
1611
1612         # Look for embedded RUTV player
1613         rutv_url = RUTVIE._extract_url(webpage)
1614         if rutv_url:
1615             return self.url_result(rutv_url, 'RUTV')
1616
1617         # Look for embedded TVC player
1618         tvc_url = TVCIE._extract_url(webpage)
1619         if tvc_url:
1620             return self.url_result(tvc_url, 'TVC')
1621
1622         # Look for embedded SportBox player
1623         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1624         if sportbox_urls:
1625             return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1626
1627         # Look for embedded PornHub player
1628         pornhub_url = PornHubIE._extract_url(webpage)
1629         if pornhub_url:
1630             return self.url_result(pornhub_url, 'PornHub')
1631
1632         # Look for embedded XHamster player
1633         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
1634         if xhamster_urls:
1635             return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
1636
1637         # Look for embedded TNAFlixNetwork player
1638         tnaflix_urls = TNAFlixNetworkEmbedIE._extract_urls(webpage)
1639         if tnaflix_urls:
1640             return _playlist_from_matches(tnaflix_urls, ie=TNAFlixNetworkEmbedIE.ie_key())
1641
1642         # Look for embedded Tvigle player
1643         mobj = re.search(
1644             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
1645         if mobj is not None:
1646             return self.url_result(mobj.group('url'), 'Tvigle')
1647
1648         # Look for embedded TED player
1649         mobj = re.search(
1650             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1651         if mobj is not None:
1652             return self.url_result(mobj.group('url'), 'TED')
1653
1654         # Look for embedded Ustream videos
1655         mobj = re.search(
1656             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1657         if mobj is not None:
1658             return self.url_result(mobj.group('url'), 'Ustream')
1659
1660         # Look for embedded arte.tv player
1661         mobj = re.search(
1662             r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1663             webpage)
1664         if mobj is not None:
1665             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1666
1667         # Look for embedded francetv player
1668         mobj = re.search(
1669             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
1670             webpage)
1671         if mobj is not None:
1672             return self.url_result(mobj.group('url'))
1673
1674         # Look for embedded smotri.com player
1675         smotri_url = SmotriIE._extract_url(webpage)
1676         if smotri_url:
1677             return self.url_result(smotri_url, 'Smotri')
1678
1679         # Look for embedded Myvi.ru player
1680         myvi_url = MyviIE._extract_url(webpage)
1681         if myvi_url:
1682             return self.url_result(myvi_url)
1683
1684         # Look for embedded soundcloud player
1685         mobj = re.search(
1686             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1687             webpage)
1688         if mobj is not None:
1689             url = unescapeHTML(mobj.group('url'))
1690             return self.url_result(url)
1691
1692         # Look for embedded vulture.com player
1693         mobj = re.search(
1694             r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1695             webpage)
1696         if mobj is not None:
1697             url = unescapeHTML(mobj.group('url'))
1698             return self.url_result(url, ie='Vulture')
1699
1700         # Look for embedded mtvservices player
1701         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
1702         if mtvservices_url:
1703             return self.url_result(mtvservices_url, ie='MTVServicesEmbedded')
1704
1705         # Look for embedded yahoo player
1706         mobj = re.search(
1707             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1708             webpage)
1709         if mobj is not None:
1710             return self.url_result(mobj.group('url'), 'Yahoo')
1711
1712         # Look for embedded sbs.com.au player
1713         mobj = re.search(
1714             r'''(?x)
1715             (?:
1716                 <meta\s+property="og:video"\s+content=|
1717                 <iframe[^>]+?src=
1718             )
1719             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1720             webpage)
1721         if mobj is not None:
1722             return self.url_result(mobj.group('url'), 'SBS')
1723
1724         # Look for embedded Cinchcast player
1725         mobj = re.search(
1726             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1727             webpage)
1728         if mobj is not None:
1729             return self.url_result(mobj.group('url'), 'Cinchcast')
1730
1731         mobj = re.search(
1732             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1733             webpage)
1734         if not mobj:
1735             mobj = re.search(
1736                 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
1737                 webpage)
1738         if mobj is not None:
1739             return self.url_result(mobj.group('url'), 'MLB')
1740
1741         mobj = re.search(
1742             r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1743             webpage)
1744         if mobj is not None:
1745             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1746
1747         mobj = re.search(
1748             r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1749             webpage)
1750         if mobj is not None:
1751             return self.url_result(mobj.group('url'), 'Livestream')
1752
1753         # Look for Zapiks embed
1754         mobj = re.search(
1755             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1756         if mobj is not None:
1757             return self.url_result(mobj.group('url'), 'Zapiks')
1758
1759         # Look for Kaltura embeds
1760         mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or
1761                 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))
1762         if mobj is not None:
1763             return self.url_result(smuggle_url(
1764                 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(),
1765                 {'source_url': url}), 'Kaltura')
1766
1767         # Look for Eagle.Platform embeds
1768         mobj = re.search(
1769             r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1770         if mobj is not None:
1771             return self.url_result(mobj.group('url'), 'EaglePlatform')
1772
1773         # Look for ClipYou (uses Eagle.Platform) embeds
1774         mobj = re.search(
1775             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1776         if mobj is not None:
1777             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1778
1779         # Look for Pladform embeds
1780         pladform_url = PladformIE._extract_url(webpage)
1781         if pladform_url:
1782             return self.url_result(pladform_url)
1783
1784         # Look for Videomore embeds
1785         videomore_url = VideomoreIE._extract_url(webpage)
1786         if videomore_url:
1787             return self.url_result(videomore_url)
1788
1789         # Look for Playwire embeds
1790         mobj = re.search(
1791             r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
1792         if mobj is not None:
1793             return self.url_result(mobj.group('url'))
1794
1795         # Look for 5min embeds
1796         mobj = re.search(
1797             r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1798         if mobj is not None:
1799             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1800
1801         # Look for Crooks and Liars embeds
1802         mobj = re.search(
1803             r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1804         if mobj is not None:
1805             return self.url_result(mobj.group('url'))
1806
1807         # Look for NBC Sports VPlayer embeds
1808         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1809         if nbc_sports_url:
1810             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1811
1812         # Look for Google Drive embeds
1813         google_drive_url = GoogleDriveIE._extract_url(webpage)
1814         if google_drive_url:
1815             return self.url_result(google_drive_url, 'GoogleDrive')
1816
1817         # Look for UDN embeds
1818         mobj = re.search(
1819             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage)
1820         if mobj is not None:
1821             return self.url_result(
1822                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1823
1824         # Look for Senate ISVP iframe
1825         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
1826         if senate_isvp_url:
1827             return self.url_result(senate_isvp_url, 'SenateISVP')
1828
1829         # Look for Dailymotion Cloud videos
1830         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
1831         if dmcloud_url:
1832             return self.url_result(dmcloud_url, 'DailymotionCloud')
1833
1834         # Look for OnionStudios embeds
1835         onionstudios_url = OnionStudiosIE._extract_url(webpage)
1836         if onionstudios_url:
1837             return self.url_result(onionstudios_url)
1838
1839         # Look for SnagFilms embeds
1840         snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage)
1841         if snagfilms_url:
1842             return self.url_result(snagfilms_url)
1843
1844         # Look for JWPlatform embeds
1845         jwplatform_url = JWPlatformIE._extract_url(webpage)
1846         if jwplatform_url:
1847             return self.url_result(jwplatform_url, 'JWPlatform')
1848
1849         # Look for ScreenwaveMedia embeds
1850         mobj = re.search(ScreenwaveMediaIE.EMBED_PATTERN, webpage)
1851         if mobj is not None:
1852             return self.url_result(unescapeHTML(mobj.group('url')), 'ScreenwaveMedia')
1853
1854         # Look for Digiteka embeds
1855         digiteka_url = DigitekaIE._extract_url(webpage)
1856         if digiteka_url:
1857             return self.url_result(self._proto_relative_url(digiteka_url), DigitekaIE.ie_key())
1858
1859         # Look for Limelight embeds
1860         mobj = re.search(r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})', webpage)
1861         if mobj:
1862             lm = {
1863                 'Media': 'media',
1864                 'Channel': 'channel',
1865                 'ChannelList': 'channel_list',
1866             }
1867             return self.url_result('limelight:%s:%s' % (
1868                 lm[mobj.group(1)], mobj.group(2)), 'Limelight%s' % mobj.group(1), mobj.group(2))
1869
1870         # Look for AdobeTVVideo embeds
1871         mobj = re.search(
1872             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
1873             webpage)
1874         if mobj is not None:
1875             return self.url_result(
1876                 self._proto_relative_url(unescapeHTML(mobj.group(1))),
1877                 'AdobeTVVideo')
1878
1879         def check_video(vurl):
1880             if YoutubeIE.suitable(vurl):
1881                 return True
1882             vpath = compat_urlparse.urlparse(vurl).path
1883             vext = determine_ext(vpath)
1884             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1885
1886         def filter_video(urls):
1887             return list(filter(check_video, urls))
1888
1889         # Start with something easy: JW Player in SWFObject
1890         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1891         if not found:
1892             # Look for gorilla-vid style embedding
1893             found = filter_video(re.findall(r'''(?sx)
1894                 (?:
1895                     jw_plugins|
1896                     JWPlayerOptions|
1897                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1898                 )
1899                 .*?
1900                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1901         if not found:
1902             # Broaden the search a little bit
1903             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1904         if not found:
1905             # Broaden the findall a little bit: JWPlayer JS loader
1906             found = filter_video(re.findall(
1907                 r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1908         if not found:
1909             # Flow player
1910             found = filter_video(re.findall(r'''(?xs)
1911                 flowplayer\("[^"]+",\s*
1912                     \{[^}]+?\}\s*,
1913                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1914                         ["']?url["']?\s*:\s*["']([^"']+)["']
1915             ''', webpage))
1916         if not found:
1917             # Cinerama player
1918             found = re.findall(
1919                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1920         if not found:
1921             # Try to find twitter cards info
1922             found = filter_video(re.findall(
1923                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1924         if not found:
1925             # We look for Open Graph info:
1926             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1927             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1928             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1929             if m_video_type is not None:
1930                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1931         if not found:
1932             # HTML5 video
1933             found = re.findall(r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1934         if not found:
1935             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1936             found = re.search(
1937                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1938                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1939                 webpage)
1940             if not found:
1941                 # Look also in Refresh HTTP header
1942                 refresh_header = head_response.headers.get('Refresh')
1943                 if refresh_header:
1944                     # In python 2 response HTTP headers are bytestrings
1945                     if sys.version_info < (3, 0) and isinstance(refresh_header, str):
1946                         refresh_header = refresh_header.decode('iso-8859-1')
1947                     found = re.search(REDIRECT_REGEX, refresh_header)
1948             if found:
1949                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
1950                 self.report_following_redirect(new_url)
1951                 return {
1952                     '_type': 'url',
1953                     'url': new_url,
1954                 }
1955         if not found:
1956             raise UnsupportedError(url)
1957
1958         entries = []
1959         for video_url in found:
1960             video_url = video_url.replace('\\/', '/')
1961             video_url = compat_urlparse.urljoin(url, video_url)
1962             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
1963
1964             # Sometimes, jwplayer extraction will result in a YouTube URL
1965             if YoutubeIE.suitable(video_url):
1966                 entries.append(self.url_result(video_url, 'Youtube'))
1967                 continue
1968
1969             # here's a fun little line of code for you:
1970             video_id = os.path.splitext(video_id)[0]
1971
1972             entry_info_dict = {
1973                 'id': video_id,
1974                 'uploader': video_uploader,
1975                 'title': video_title,
1976                 'age_limit': age_limit,
1977             }
1978
1979             ext = determine_ext(video_url)
1980             if ext == 'smil':
1981                 entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
1982             elif ext == 'xspf':
1983                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
1984             elif ext == 'm3u8':
1985                 entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
1986             elif ext == 'mpd':
1987                 entry_info_dict['formats'] = self._extract_mpd_formats(video_url, video_id)
1988             else:
1989                 entry_info_dict['url'] = video_url
1990
1991             entries.append(entry_info_dict)
1992
1993         if len(entries) == 1:
1994             return entries[0]
1995         else:
1996             for num, e in enumerate(entries, start=1):
1997                 # 'url' results don't have a title
1998                 if e.get('title') is not None:
1999                     e['title'] = '%s (%d)' % (e['title'], num)
2000             return {
2001                 '_type': 'playlist',
2002                 'entries': entries,
2003             }