_ 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 from .instagram import InstagramIE
  63 from .liveleak import LiveLeakIE
  64
  65
  66 class GenericIE(InfoExtractor):
  67     IE_DESC = 'Generic downloader that works on some sites'
  68     _VALID_URL = r'.*'
  69     IE_NAME = 'generic'
  70     _TESTS = [
  71         # Direct link to a video
  72         {
  73             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  74             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  75             'info_dict': {
  76                 'id': 'trailer',
  77                 'ext': 'mp4',
  78                 'title': 'trailer',
  79                 'upload_date': '20100513',
  80             }
  81         },
  82         # Direct link to media delivered compressed (until Accept-Encoding is *)
  83         {
  84             'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
  85             'md5': '128c42e68b13950268b648275386fc74',
  86             'info_dict': {
  87                 'id': 'FictionJunction-Parallel_Hearts',
  88                 'ext': 'flac',
  89                 'title': 'FictionJunction-Parallel_Hearts',
  90                 'upload_date': '20140522',
  91             },
  92             'expected_warnings': [
  93                 'URL could be a direct video link, returning it as such.'
  94             ]
  95         },
  96         # Direct download with broken HEAD
  97         {
  98             'url': 'http://ai-radio.org:8000/radio.opus',
  99             'info_dict': {
 100                 'id': 'radio',
 101                 'ext': 'opus',
 102                 'title': 'radio',
 103             },
 104             'params': {
 105                 'skip_download': True,  # infinite live stream
 106             },
 107             'expected_warnings': [
 108                 r'501.*Not Implemented'
 109             ],
 110         },
 111         # Direct link with incorrect MIME type
 112         {
 113             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 114             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 115             'info_dict': {
 116                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 117                 'id': '5_Lennart_Poettering_-_Systemd',
 118                 'ext': 'webm',
 119                 'title': '5_Lennart_Poettering_-_Systemd',
 120                 'upload_date': '20141120',
 121             },
 122             'expected_warnings': [
 123                 'URL could be a direct video link, returning it as such.'
 124             ]
 125         },
 126         # RSS feed
 127         {
 128             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 129             'info_dict': {
 130                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 131                 'title': 'Zero Punctuation',
 132                 'description': 're:.*groundbreaking video review series.*'
 133             },
 134             'playlist_mincount': 11,
 135         },
 136         # RSS feed with enclosure
 137         {
 138             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 139             'info_dict': {
 140                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 141                 'ext': 'm4v',
 142                 'upload_date': '20150228',
 143                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 144             }
 145         },
 146         # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
 147         {
 148             'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
 149             'info_dict': {
 150                 'id': 'smil',
 151                 'ext': 'mp4',
 152                 'title': 'Automatics, robotics and biocybernetics',
 153                 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
 154                 'upload_date': '20130627',
 155                 'formats': 'mincount:16',
 156                 'subtitles': 'mincount:1',
 157             },
 158             'params': {
 159                 'force_generic_extractor': True,
 160                 'skip_download': True,
 161             },
 162         },
 163         # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html
 164         {
 165             'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil',
 166             'info_dict': {
 167                 'id': 'hds',
 168                 'ext': 'flv',
 169                 'title': 'hds',
 170                 'formats': 'mincount:1',
 171             },
 172             'params': {
 173                 'skip_download': True,
 174             },
 175         },
 176         # SMIL from https://www.restudy.dk/video/play/id/1637
 177         {
 178             'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml',
 179             'info_dict': {
 180                 'id': 'video_1637',
 181                 'ext': 'flv',
 182                 'title': 'video_1637',
 183                 'formats': 'mincount:3',
 184             },
 185             'params': {
 186                 'skip_download': True,
 187             },
 188         },
 189         # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm
 190         {
 191             'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil',
 192             'info_dict': {
 193                 'id': 'smil-service',
 194                 'ext': 'flv',
 195                 'title': 'smil-service',
 196                 'formats': 'mincount:1',
 197             },
 198             'params': {
 199                 'skip_download': True,
 200             },
 201         },
 202         # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370
 203         {
 204             'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil',
 205             'info_dict': {
 206                 'id': '4719370',
 207                 'ext': 'mp4',
 208                 'title': '571de1fd-47bc-48db-abf9-238872a58d1f',
 209                 'formats': 'mincount:3',
 210             },
 211             'params': {
 212                 'skip_download': True,
 213             },
 214         },
 215         # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html
 216         {
 217             'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf',
 218             'info_dict': {
 219                 'id': 'mZlp2ctYIUEB',
 220                 'ext': 'mp4',
 221                 'title': 'Tikibad ontruimd wegens brand',
 222                 'description': 'md5:05ca046ff47b931f9b04855015e163a4',
 223                 'thumbnail': 're:^https?://.*\.jpg$',
 224                 'duration': 33,
 225             },
 226             'params': {
 227                 'skip_download': True,
 228             },
 229         },
 230         # MPD from http://dash-mse-test.appspot.com/media.html
 231         {
 232             'url': 'http://yt-dash-mse-test.commondatastorage.googleapis.com/media/car-20120827-manifest.mpd',
 233             'md5': '4b57baab2e30d6eb3a6a09f0ba57ef53',
 234             'info_dict': {
 235                 'id': 'car-20120827-manifest',
 236                 'ext': 'mp4',
 237                 'title': 'car-20120827-manifest',
 238                 'formats': 'mincount:9',
 239             },
 240             'params': {
 241                 'format': 'bestvideo',
 242             },
 243         },
 244         # m3u8 served with Content-Type: audio/x-mpegURL; charset=utf-8
 245         {
 246             'url': 'http://once.unicornmedia.com/now/master/playlist/bb0b18ba-64f5-4b1b-a29f-0ac252f06b68/77a785f3-5188-4806-b788-0893a61634ed/93677179-2d99-4ef4-9e17-fe70d49abfbf/content.m3u8',
 247             'info_dict': {
 248                 'id': 'content',
 249                 'ext': 'mp4',
 250                 'title': 'content',
 251                 'formats': 'mincount:8',
 252             },
 253             'params': {
 254                 # m3u8 downloads
 255                 'skip_download': True,
 256             }
 257         },
 258         # m3u8 served with Content-Type: text/plain
 259         {
 260             'url': 'http://www.nacentapps.com/m3u8/index.m3u8',
 261             'info_dict': {
 262                 'id': 'index',
 263                 'ext': 'mp4',
 264                 'title': 'index',
 265                 'upload_date': '20140720',
 266                 'formats': 'mincount:11',
 267             },
 268             'params': {
 269                 # m3u8 downloads
 270                 'skip_download': True,
 271             }
 272         },
 273         # google redirect
 274         {
 275             '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',
 276             'info_dict': {
 277                 'id': 'cmQHVoWB5FY',
 278                 'ext': 'mp4',
 279                 'upload_date': '20130224',
 280                 'uploader_id': 'TheVerge',
 281                 'description': 're:^Chris Ziegler takes a look at the\.*',
 282                 'uploader': 'The Verge',
 283                 'title': 'First Firefox OS phones side-by-side',
 284             },
 285             'params': {
 286                 'skip_download': False,
 287             }
 288         },
 289         {
 290             # redirect in Refresh HTTP header
 291             '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',
 292             'info_dict': {
 293                 'id': 'pO8h3EaFRdo',
 294                 'ext': 'mp4',
 295                 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
 296                 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5',
 297                 'upload_date': '20150917',
 298                 'uploader_id': 'brtvofficial',
 299                 'uploader': 'Boiler Room',
 300             },
 301             'params': {
 302                 'skip_download': False,
 303             },
 304         },
 305         {
 306             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
 307             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
 308             'info_dict': {
 309                 'id': '13601338388002',
 310                 'ext': 'mp4',
 311                 'uploader': 'www.hodiho.fr',
 312                 'title': 'R\u00e9gis plante sa Jeep',
 313             }
 314         },
 315         # bandcamp page with custom domain
 316         {
 317             'add_ie': ['Bandcamp'],
 318             'url': 'http://bronyrock.com/track/the-pony-mash',
 319             'info_dict': {
 320                 'id': '3235767654',
 321                 'ext': 'mp3',
 322                 'title': 'The Pony Mash',
 323                 'uploader': 'M_Pallante',
 324             },
 325             'skip': 'There is a limit of 200 free downloads / month for the test song',
 326         },
 327         # embedded brightcove video
 328         # it also tests brightcove videos that need to set the 'Referer' in the
 329         # http requests
 330         {
 331             'add_ie': ['BrightcoveLegacy'],
 332             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
 333             'info_dict': {
 334                 'id': '2765128793001',
 335                 'ext': 'mp4',
 336                 'title': 'Le cours de bourse : l’analyse technique',
 337                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
 338                 'uploader': 'BFM BUSINESS',
 339             },
 340             'params': {
 341                 'skip_download': True,
 342             },
 343         },
 344         {
 345             # https://github.com/rg3/youtube-dl/issues/2253
 346             'url': 'http://bcove.me/i6nfkrc3',
 347             'md5': '0ba9446db037002366bab3b3eb30c88c',
 348             'info_dict': {
 349                 'id': '3101154703001',
 350                 'ext': 'mp4',
 351                 'title': 'Still no power',
 352                 'uploader': 'thestar.com',
 353                 '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.',
 354             },
 355             'add_ie': ['BrightcoveLegacy'],
 356         },
 357         {
 358             'url': 'http://www.championat.com/video/football/v/87/87499.html',
 359             'md5': 'fb973ecf6e4a78a67453647444222983',
 360             'info_dict': {
 361                 'id': '3414141473001',
 362                 'ext': 'mp4',
 363                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 364                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 365                 'uploader': 'Championat',
 366             },
 367         },
 368         {
 369             # https://github.com/rg3/youtube-dl/issues/3541
 370             'add_ie': ['BrightcoveLegacy'],
 371             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 372             'info_dict': {
 373                 'id': '3866516442001',
 374                 'ext': 'mp4',
 375                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 376                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 377                 'uploader': 'SBS Broadcasting',
 378             },
 379             'skip': 'Restricted to Netherlands',
 380             'params': {
 381                 'skip_download': True,  # m3u8 download
 382             },
 383         },
 384         # ooyala video
 385         {
 386             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 387             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 388             'info_dict': {
 389                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 390                 'ext': 'mp4',
 391                 'title': '2cc213299525360.mov',  # that's what we get
 392                 'duration': 238.231,
 393             },
 394             'add_ie': ['Ooyala'],
 395         },
 396         {
 397             # ooyala video embedded with http://player.ooyala.com/iframe.js
 398             'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/',
 399             'info_dict': {
 400                 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB',
 401                 'ext': 'mp4',
 402                 'title': '"Steve Jobs: Man in the Machine" trailer',
 403                 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
 404                 'duration': 135.427,
 405             },
 406             'params': {
 407                 'skip_download': True,
 408             },
 409         },
 410         # embed.ly video
 411         {
 412             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 413             'info_dict': {
 414                 'id': '9ODmcdjQcHQ',
 415                 'ext': 'mp4',
 416                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 417                 'upload_date': '20140225',
 418                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 419                 'uploader': 'Tested',
 420                 'uploader_id': 'testedcom',
 421             },
 422             # No need to test YoutubeIE here
 423             'params': {
 424                 'skip_download': True,
 425             },
 426         },
 427         # funnyordie embed
 428         {
 429             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 430             'info_dict': {
 431                 'id': '18e820ec3f',
 432                 'ext': 'mp4',
 433                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 434                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 435             },
 436         },
 437         # RUTV embed
 438         {
 439             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 440             'info_dict': {
 441                 'id': '776940',
 442                 'ext': 'mp4',
 443                 'title': 'Охотское море стало целиком российским',
 444                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 445             },
 446             'params': {
 447                 # m3u8 download
 448                 'skip_download': True,
 449             },
 450         },
 451         # TVC embed
 452         {
 453             '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/',
 454             'info_dict': {
 455                 'id': '55304',
 456                 'ext': 'mp4',
 457                 'title': 'Дошкольное воспитание',
 458             },
 459         },
 460         # SportBox embed
 461         {
 462             'url': 'http://www.vestifinance.ru/articles/25753',
 463             'info_dict': {
 464                 'id': '25753',
 465                 'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
 466             },
 467             'playlist': [{
 468                 'info_dict': {
 469                     'id': '370908',
 470                     'title': 'Госзаказ. День 3',
 471                     'ext': 'mp4',
 472                 }
 473             }, {
 474                 'info_dict': {
 475                     'id': '370905',
 476                     'title': 'Госзаказ. День 2',
 477                     'ext': 'mp4',
 478                 }
 479             }, {
 480                 'info_dict': {
 481                     'id': '370902',
 482                     'title': 'Госзаказ. День 1',
 483                     'ext': 'mp4',
 484                 }
 485             }],
 486             'params': {
 487                 # m3u8 download
 488                 'skip_download': True,
 489             },
 490         },
 491         # Myvi.ru embed
 492         {
 493             'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
 494             'info_dict': {
 495                 'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
 496                 'ext': 'mp4',
 497                 'title': 'Ужастики, русский трейлер (2015)',
 498                 'thumbnail': 're:^https?://.*\.jpg$',
 499                 'duration': 153,
 500             }
 501         },
 502         # XHamster embed
 503         {
 504             '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',
 505             'info_dict': {
 506                 'id': 'showthread',
 507                 'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
 508             },
 509             'playlist_mincount': 7,
 510         },
 511         # Embedded TED video
 512         {
 513             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 514             'md5': '65fdff94098e4a607385a60c5177c638',
 515             'info_dict': {
 516                 'id': '1969',
 517                 'ext': 'mp4',
 518                 'title': 'Hidden miracles of the natural world',
 519                 'uploader': 'Louie Schwartzberg',
 520                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 521             }
 522         },
 523         # Embedded Ustream video
 524         {
 525             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 526             'md5': '27b99cdb639c9b12a79bca876a073417',
 527             'info_dict': {
 528                 'id': '45734260',
 529                 'ext': 'flv',
 530                 'uploader': 'AU SPA:  The NSA and Privacy',
 531                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 532             }
 533         },
 534         # nowvideo embed hidden behind percent encoding
 535         {
 536             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 537             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 538             'info_dict': {
 539                 'id': '06e53103ca9aa',
 540                 'ext': 'flv',
 541                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 542                 'description': 'No description',
 543             },
 544         },
 545         # arte embed
 546         {
 547             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 548             'md5': '7653032cbb25bf6c80d80f217055fa43',
 549             'info_dict': {
 550                 'id': '048195-004_PLUS7-F',
 551                 'ext': 'flv',
 552                 'title': 'X:enius',
 553                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 554                 'upload_date': '20140320',
 555             },
 556             'params': {
 557                 'skip_download': 'Requires rtmpdump'
 558             }
 559         },
 560         # francetv embed
 561         {
 562             'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
 563             'info_dict': {
 564                 'id': 'EV_30231',
 565                 'ext': 'mp4',
 566                 'title': 'Alcaline, le concert avec Calogero',
 567                 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
 568                 'upload_date': '20150226',
 569                 'timestamp': 1424989860,
 570                 'duration': 5400,
 571             },
 572             'params': {
 573                 # m3u8 downloads
 574                 'skip_download': True,
 575             },
 576             'expected_warnings': [
 577                 'Forbidden'
 578             ]
 579         },
 580         # Condé Nast embed
 581         {
 582             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 583             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 584             'info_dict': {
 585                 'id': '53501be369702d3275860000',
 586                 'ext': 'mp4',
 587                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 588             }
 589         },
 590         # Dailymotion embed
 591         {
 592             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 593             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 594             'info_dict': {
 595                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 596                 'ext': 'mp4',
 597                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 598                 'uploader': 'Spi0n',
 599             },
 600             'add_ie': ['Dailymotion'],
 601         },
 602         # YouTube embed
 603         {
 604             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 605             'info_dict': {
 606                 'id': 'FXRb4ykk4S0',
 607                 'ext': 'mp4',
 608                 'title': 'The NBL Auction 2014',
 609                 'uploader': 'BADMINTON England',
 610                 'uploader_id': 'BADMINTONEvents',
 611                 'upload_date': '20140603',
 612                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 613             },
 614             'add_ie': ['Youtube'],
 615             'params': {
 616                 'skip_download': True,
 617             }
 618         },
 619         # MTVSercices embed
 620         {
 621             'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
 622             'md5': '35727f82f58c76d996fc188f9755b0d5',
 623             'info_dict': {
 624                 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
 625                 'ext': 'mp4',
 626                 'title': 'Review',
 627                 'description': 'Mario\'s life in the fast lane has never looked so good.',
 628             },
 629         },
 630         # YouTube embed via <data-embed-url="">
 631         {
 632             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 633             'info_dict': {
 634                 'id': '4vAffPZIT44',
 635                 'ext': 'mp4',
 636                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 637                 'uploader': 'Gameloft',
 638                 'uploader_id': 'gameloft',
 639                 'upload_date': '20140828',
 640                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 641             },
 642             'params': {
 643                 'skip_download': True,
 644             }
 645         },
 646         # Camtasia studio
 647         {
 648             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 649             'playlist': [{
 650                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 651                 'info_dict': {
 652                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 653                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 654                     'ext': 'flv',
 655                     'duration': 2235.90,
 656                 }
 657             }, {
 658                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 659                 'info_dict': {
 660                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 661                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 662                     'ext': 'flv',
 663                     'duration': 2235.93,
 664                 }
 665             }],
 666             'info_dict': {
 667                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 668             }
 669         },
 670         # Flowplayer
 671         {
 672             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 673             'md5': '9d65602bf31c6e20014319c7d07fba27',
 674             'info_dict': {
 675                 'id': '5123ea6d5e5a7',
 676                 'ext': 'mp4',
 677                 'age_limit': 18,
 678                 'uploader': 'www.handjobhub.com',
 679                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 680             }
 681         },
 682         # Multiple brightcove videos
 683         # https://github.com/rg3/youtube-dl/issues/2283
 684         {
 685             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 686             'info_dict': {
 687                 'id': 'always-never',
 688                 'title': 'Always / Never - The New Yorker',
 689             },
 690             'playlist_count': 3,
 691             'params': {
 692                 'extract_flat': False,
 693                 'skip_download': True,
 694             }
 695         },
 696         # MLB embed
 697         {
 698             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 699             'md5': '96f09a37e44da40dd083e12d9a683327',
 700             'info_dict': {
 701                 'id': '33322633',
 702                 'ext': 'mp4',
 703                 'title': 'Ump changes call to ball',
 704                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 705                 'duration': 48,
 706                 'timestamp': 1401537900,
 707                 'upload_date': '20140531',
 708                 'thumbnail': 're:^https?://.*\.jpg$',
 709             },
 710         },
 711         # Wistia embed
 712         {
 713             'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 714             'md5': '8788b683c777a5cf25621eaf286d0c23',
 715             'info_dict': {
 716                 'id': '1cfaf6b7ea',
 717                 'ext': 'mov',
 718                 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
 719                 'duration': 643.0,
 720                 'filesize': 182808282,
 721                 'uploader': 'education-portal.com',
 722             },
 723         },
 724         {
 725             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 726             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 727             'info_dict': {
 728                 'id': 'uxjb0lwrcz',
 729                 'ext': 'mp4',
 730                 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
 731                 'duration': 1715.0,
 732                 'uploader': 'thoughtworks.wistia.com',
 733             },
 734         },
 735         # Soundcloud embed
 736         {
 737             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 738             'info_dict': {
 739                 'id': '174391317',
 740                 'ext': 'mp3',
 741                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 742                 'uploader': 'Sophos Security',
 743                 'title': 'Chet Chat 171 - Oct 29, 2014',
 744                 'upload_date': '20141029',
 745             }
 746         },
 747         # Livestream embed
 748         {
 749             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 750             'info_dict': {
 751                 'id': '67864563',
 752                 'ext': 'flv',
 753                 'upload_date': '20141112',
 754                 'title': 'Rosetta #CometLanding webcast HL 10',
 755             }
 756         },
 757         # LazyYT
 758         {
 759             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
 760             'info_dict': {
 761                 'id': '1986',
 762                 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
 763             },
 764             'playlist_mincount': 2,
 765         },
 766         # Cinchcast embed
 767         {
 768             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
 769             'info_dict': {
 770                 'id': '7141703',
 771                 'ext': 'mp3',
 772                 'upload_date': '20141126',
 773                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
 774             }
 775         },
 776         # Cinerama player
 777         {
 778             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
 779             'info_dict': {
 780                 'id': '730m_DandD_1901_512k',
 781                 'ext': 'mp4',
 782                 'uploader': 'www.abc.net.au',
 783                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
 784             }
 785         },
 786         # embedded viddler video
 787         {
 788             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
 789             'info_dict': {
 790                 'id': '4d03aad9',
 791                 'ext': 'mp4',
 792                 'uploader': 'deadspin',
 793                 'title': 'WALL-TO-GORTAT',
 794                 'timestamp': 1422285291,
 795                 'upload_date': '20150126',
 796             },
 797             'add_ie': ['Viddler'],
 798         },
 799         # Libsyn embed
 800         {
 801             'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
 802             'info_dict': {
 803                 'id': '3377616',
 804                 'ext': 'mp3',
 805                 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
 806                 'description': 'md5:601cb790edd05908957dae8aaa866465',
 807                 'upload_date': '20150220',
 808             },
 809         },
 810         # jwplayer YouTube
 811         {
 812             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
 813             'info_dict': {
 814                 'id': 'Mrj4DVp2zeA',
 815                 'ext': 'mp4',
 816                 'upload_date': '20150212',
 817                 'uploader': 'The National Archives UK',
 818                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
 819                 'uploader_id': 'NationalArchives08',
 820                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
 821             },
 822         },
 823         # rtl.nl embed
 824         {
 825             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
 826             'playlist_mincount': 5,
 827             'info_dict': {
 828                 'id': 'aanslagen-kopenhagen',
 829                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
 830             }
 831         },
 832         # Zapiks embed
 833         {
 834             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
 835             'info_dict': {
 836                 'id': '118046',
 837                 'ext': 'mp4',
 838                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
 839             }
 840         },
 841         # Kaltura embed
 842         {
 843             'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
 844             'info_dict': {
 845                 'id': '1_eergr3h1',
 846                 'ext': 'mp4',
 847                 'upload_date': '20150226',
 848                 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
 849                 'timestamp': int,
 850                 'title': 'John Carlson Postgame 2/25/15',
 851             },
 852         },
 853         # Kaltura embed (different embed code)
 854         {
 855             'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
 856             'info_dict': {
 857                 'id': '1_a52wc67y',
 858                 'ext': 'flv',
 859                 'upload_date': '20150127',
 860                 'uploader_id': 'PremierMedia',
 861                 'timestamp': int,
 862                 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
 863             },
 864         },
 865         # Kaltura embed protected with referrer
 866         {
 867             'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero',
 868             'info_dict': {
 869                 'id': '1_g4fbemnq',
 870                 'ext': 'mp4',
 871                 'title': 'Violetta - Achter De Schermen - Ruggero',
 872                 'description': 'Achter de schermen met Ruggero',
 873                 'timestamp': 1435133761,
 874                 'upload_date': '20150624',
 875                 'uploader_id': 'echojecka',
 876             },
 877         },
 878         # Eagle.Platform embed (generic URL)
 879         {
 880             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
 881             'info_dict': {
 882                 'id': '227304',
 883                 'ext': 'mp4',
 884                 'title': 'Навальный вышел на свободу',
 885                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
 886                 'thumbnail': 're:^https?://.*\.jpg$',
 887                 'duration': 87,
 888                 'view_count': int,
 889                 'age_limit': 0,
 890             },
 891         },
 892         # ClipYou (Eagle.Platform) embed (custom URL)
 893         {
 894             'url': 'http://muz-tv.ru/play/7129/',
 895             'info_dict': {
 896                 'id': '12820',
 897                 'ext': 'mp4',
 898                 'title': "'O Sole Mio",
 899                 'thumbnail': 're:^https?://.*\.jpg$',
 900                 'duration': 216,
 901                 'view_count': int,
 902             },
 903         },
 904         # Pladform embed
 905         {
 906             'url': 'http://muz-tv.ru/kinozal/view/7400/',
 907             'info_dict': {
 908                 'id': '100183293',
 909                 'ext': 'mp4',
 910                 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
 911                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
 912                 'thumbnail': 're:^https?://.*\.jpg$',
 913                 'duration': 694,
 914                 'age_limit': 0,
 915             },
 916         },
 917         # Playwire embed
 918         {
 919             'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
 920             'info_dict': {
 921                 'id': '3519514',
 922                 'ext': 'mp4',
 923                 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
 924                 'thumbnail': 're:^https?://.*\.png$',
 925                 'duration': 45.115,
 926             },
 927         },
 928         # 5min embed
 929         {
 930             'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
 931             'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
 932             'info_dict': {
 933                 'id': '518726732',
 934                 'ext': 'mp4',
 935                 'title': 'Facebook Creates "On This Day" | Crunch Report',
 936             },
 937         },
 938         # SVT embed
 939         {
 940             'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 941             'info_dict': {
 942                 'id': '2900353',
 943                 'ext': 'flv',
 944                 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
 945                 'duration': 27,
 946                 'age_limit': 0,
 947             },
 948         },
 949         # Crooks and Liars embed
 950         {
 951             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
 952             'info_dict': {
 953                 'id': '8RUoRhRi',
 954                 'ext': 'mp4',
 955                 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
 956                 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
 957                 'timestamp': 1428207000,
 958                 'upload_date': '20150405',
 959                 'uploader': 'Heather',
 960             },
 961         },
 962         # Crooks and Liars external embed
 963         {
 964             'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
 965             'info_dict': {
 966                 'id': 'MTE3MjUtMzQ2MzA',
 967                 'ext': 'mp4',
 968                 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
 969                 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
 970                 'timestamp': 1265032391,
 971                 'upload_date': '20100201',
 972                 'uploader': 'Heather',
 973             },
 974         },
 975         # NBC Sports vplayer embed
 976         {
 977             'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
 978             'info_dict': {
 979                 'id': 'ln7x1qSThw4k',
 980                 'ext': 'flv',
 981                 'title': "PFT Live: New leader in the 'new-look' defense",
 982                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
 983             },
 984         },
 985         # UDN embed
 986         {
 987             'url': 'http://www.udn.com/news/story/7314/822787',
 988             'md5': 'fd2060e988c326991037b9aff9df21a6',
 989             'info_dict': {
 990                 'id': '300346',
 991                 'ext': 'mp4',
 992                 'title': '中一中男師變性 全校師生力挺',
 993                 'thumbnail': 're:^https?://.*\.jpg$',
 994             }
 995         },
 996         # Ooyala embed
 997         {
 998             'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
 999             'info_dict': {
1000                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
1001                 'ext': 'mp4',
1002                 'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.',
1003                 'title': 'This is what separates the Excel masters from the wannabes',
1004                 'duration': 191.933,
1005             },
1006             'params': {
1007                 # m3u8 downloads
1008                 'skip_download': True,
1009             }
1010         },
1011         # Contains a SMIL manifest
1012         {
1013             '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',
1014             'info_dict': {
1015                 'id': 'file',
1016                 'ext': 'flv',
1017                 'title': '+ Football: Lottery Champions League Europe',
1018                 'uploader': 'www.telewebion.com',
1019             },
1020             'params': {
1021                 # rtmpe downloads
1022                 'skip_download': True,
1023             }
1024         },
1025         # Brightcove URL in single quotes
1026         {
1027             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
1028             'md5': '4ae374f1f8b91c889c4b9203c8c752af',
1029             'info_dict': {
1030                 'id': '4255764656001',
1031                 'ext': 'mp4',
1032                 'title': 'SN Presents: Russell Martin, World Citizen',
1033                 '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.',
1034                 'uploader': 'Rogers Sportsnet',
1035             },
1036         },
1037         # Dailymotion Cloud video
1038         {
1039             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
1040             'md5': '49444254273501a64675a7e68c502681',
1041             'info_dict': {
1042                 'id': '5585de919473990de4bee11b',
1043                 'ext': 'mp4',
1044                 'title': 'Le débat',
1045                 'thumbnail': 're:^https?://.*\.jpe?g$',
1046             }
1047         },
1048         # OnionStudios embed
1049         {
1050             'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
1051             'info_dict': {
1052                 'id': '2855',
1053                 'ext': 'mp4',
1054                 'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
1055                 'thumbnail': 're:^https?://.*\.jpe?g$',
1056                 'uploader': 'ClickHole',
1057                 'uploader_id': 'clickhole',
1058             }
1059         },
1060         # SnagFilms embed
1061         {
1062             'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
1063             'info_dict': {
1064                 'id': '74849a00-85a9-11e1-9660-123139220831',
1065                 'ext': 'mp4',
1066                 'title': '#whilewewatch',
1067             }
1068         },
1069         # AdobeTVVideo embed
1070         {
1071             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
1072             'md5': '43662b577c018ad707a63766462b1e87',
1073             'info_dict': {
1074                 'id': '2456',
1075                 'ext': 'mp4',
1076                 'title': 'New experience with Acrobat DC',
1077                 'description': 'New experience with Acrobat DC',
1078                 'duration': 248.667,
1079             },
1080         },
1081         # ScreenwaveMedia embed
1082         {
1083             'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1',
1084             'md5': '24ace5baba0d35d55c6810b51f34e9e0',
1085             'info_dict': {
1086                 'id': 'cinemasnob-55d26273809dd',
1087                 'ext': 'mp4',
1088                 'title': 'cinemasnob',
1089             },
1090         },
1091         # BrightcoveInPageEmbed embed
1092         {
1093             'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
1094             'info_dict': {
1095                 'id': '4238694884001',
1096                 'ext': 'flv',
1097                 'title': 'Tabletop: Dread, Last Thoughts',
1098                 'description': 'Tabletop: Dread, Last Thoughts',
1099                 'duration': 51690,
1100             },
1101         },
1102         # JWPlayer with M3U8
1103         {
1104             'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video',
1105             'info_dict': {
1106                 'id': 'playlist',
1107                 'ext': 'mp4',
1108                 'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ',
1109                 'uploader': 'ren.tv',
1110             },
1111             'params': {
1112                 # m3u8 downloads
1113                 'skip_download': True,
1114             }
1115         },
1116         # Brightcove embed, with no valid 'renditions' but valid 'IOSRenditions'
1117         # This video can't be played in browsers if Flash disabled and UA set to iPhone, which is actually a false alarm
1118         {
1119             'url': 'https://dl.dropboxusercontent.com/u/29092637/interview.html',
1120             'info_dict': {
1121                 'id': '4785848093001',
1122                 'ext': 'mp4',
1123                 'title': 'The Cardinal Pell Interview',
1124                 'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ',
1125                 'uploader': 'GlobeCast Australia - GlobeStream',
1126             },
1127             'params': {
1128                 # m3u8 downloads
1129                 'skip_download': True,
1130             },
1131         },
1132         # Another form of arte.tv embed
1133         {
1134             'url': 'http://www.tv-replay.fr/redirection/09-04-16/arte-reportage-arte-11508975.html',
1135             'md5': '850bfe45417ddf221288c88a0cffe2e2',
1136             'info_dict': {
1137                 'id': '030273-562_PLUS7-F',
1138                 'ext': 'mp4',
1139                 'title': 'ARTE Reportage - Nulle part, en France',
1140                 'description': 'md5:e3a0e8868ed7303ed509b9e3af2b870d',
1141                 'upload_date': '20160409',
1142             },
1143         },
1144         # LiveLeak embed
1145         {
1146             'url': 'http://www.wykop.pl/link/3088787/',
1147             'md5': 'ace83b9ed19b21f68e1b50e844fdf95d',
1148             'info_dict': {
1149                 'id': '874_1459135191',
1150                 'ext': 'mp4',
1151                 'title': 'Man shows poor quality of new apartment building',
1152                 'description': 'The wall is like a sand pile.',
1153                 'uploader': 'Lake8737',
1154             }
1155         },
1156     ]
1157
1158     def report_following_redirect(self, new_url):
1159         """Report information extraction."""
1160         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
1161
1162     def _extract_rss(self, url, video_id, doc):
1163         playlist_title = doc.find('./channel/title').text
1164         playlist_desc_el = doc.find('./channel/description')
1165         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
1166
1167         entries = []
1168         for it in doc.findall('./channel/item'):
1169             next_url = xpath_text(it, 'link', fatal=False)
1170             if not next_url:
1171                 enclosure_nodes = it.findall('./enclosure')
1172                 for e in enclosure_nodes:
1173                     next_url = e.attrib.get('url')
1174                     if next_url:
1175                         break
1176
1177             if not next_url:
1178                 continue
1179
1180             entries.append({
1181                 '_type': 'url',
1182                 'url': next_url,
1183                 'title': it.find('title').text,
1184             })
1185
1186         return {
1187             '_type': 'playlist',
1188             'id': url,
1189             'title': playlist_title,
1190             'description': playlist_desc,
1191             'entries': entries,
1192         }
1193
1194     def _extract_camtasia(self, url, video_id, webpage):
1195         """ Returns None if no camtasia video can be found. """
1196
1197         camtasia_cfg = self._search_regex(
1198             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
1199             webpage, 'camtasia configuration file', default=None)
1200         if camtasia_cfg is None:
1201             return None
1202
1203         title = self._html_search_meta('DC.title', webpage, fatal=True)
1204
1205         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
1206         camtasia_cfg = self._download_xml(
1207             camtasia_url, video_id,
1208             note='Downloading camtasia configuration',
1209             errnote='Failed to download camtasia configuration')
1210         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
1211
1212         entries = []
1213         for n in fileset_node.getchildren():
1214             url_n = n.find('./uri')
1215             if url_n is None:
1216                 continue
1217
1218             entries.append({
1219                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
1220                 'title': '%s - %s' % (title, n.tag),
1221                 'url': compat_urlparse.urljoin(url, url_n.text),
1222                 'duration': float_or_none(n.find('./duration').text),
1223             })
1224
1225         return {
1226             '_type': 'playlist',
1227             'entries': entries,
1228             'title': title,
1229         }
1230
1231     def _real_extract(self, url):
1232         if url.startswith('//'):
1233             return {
1234                 '_type': 'url',
1235                 'url': self.http_scheme() + url,
1236             }
1237
1238         parsed_url = compat_urlparse.urlparse(url)
1239         if not parsed_url.scheme:
1240             default_search = self._downloader.params.get('default_search')
1241             if default_search is None:
1242                 default_search = 'fixup_error'
1243
1244             if default_search in ('auto', 'auto_warning', 'fixup_error'):
1245                 if '/' in url:
1246                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
1247                     return self.url_result('http://' + url)
1248                 elif default_search != 'fixup_error':
1249                     if default_search == 'auto_warning':
1250                         if re.match(r'^(?:url|URL)$', url):
1251                             raise ExtractorError(
1252                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
1253                                 expected=True)
1254                         else:
1255                             self._downloader.report_warning(
1256                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
1257                     return self.url_result('ytsearch:' + url)
1258
1259             if default_search in ('error', 'fixup_error'):
1260                 raise ExtractorError(
1261                     '%r is not a valid URL. '
1262                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
1263                     % (url, url), expected=True)
1264             else:
1265                 if ':' not in default_search:
1266                     default_search += ':'
1267                 return self.url_result(default_search + url)
1268
1269         url, smuggled_data = unsmuggle_url(url)
1270         force_videoid = None
1271         is_intentional = smuggled_data and smuggled_data.get('to_generic')
1272         if smuggled_data and 'force_videoid' in smuggled_data:
1273             force_videoid = smuggled_data['force_videoid']
1274             video_id = force_videoid
1275         else:
1276             video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
1277
1278         self.to_screen('%s: Requesting header' % video_id)
1279
1280         head_req = HEADRequest(url)
1281         head_response = self._request_webpage(
1282             head_req, video_id,
1283             note=False, errnote='Could not send HEAD request to %s' % url,
1284             fatal=False)
1285
1286         if head_response is not False:
1287             # Check for redirect
1288             new_url = head_response.geturl()
1289             if url != new_url:
1290                 self.report_following_redirect(new_url)
1291                 if force_videoid:
1292                     new_url = smuggle_url(
1293                         new_url, {'force_videoid': force_videoid})
1294                 return self.url_result(new_url)
1295
1296         full_response = None
1297         if head_response is False:
1298             request = sanitized_Request(url)
1299             request.add_header('Accept-Encoding', '*')
1300             full_response = self._request_webpage(request, video_id)
1301             head_response = full_response
1302
1303         info_dict = {
1304             'id': video_id,
1305             'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1306             'upload_date': unified_strdate(head_response.headers.get('Last-Modified'))
1307         }
1308
1309         # Check for direct link to a video
1310         content_type = head_response.headers.get('Content-Type', '').lower()
1311         m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>[^;\s]+)', content_type)
1312         if m:
1313             format_id = m.group('format_id')
1314             if format_id.endswith('mpegurl'):
1315                 formats = self._extract_m3u8_formats(url, video_id, 'mp4')
1316             elif format_id == 'f4m':
1317                 formats = self._extract_f4m_formats(url, video_id)
1318             else:
1319                 formats = [{
1320                     'format_id': m.group('format_id'),
1321                     'url': url,
1322                     'vcodec': 'none' if m.group('type') == 'audio' else None
1323                 }]
1324                 info_dict['direct'] = True
1325             self._sort_formats(formats)
1326             info_dict['formats'] = formats
1327             return info_dict
1328
1329         if not self._downloader.params.get('test', False) and not is_intentional:
1330             force = self._downloader.params.get('force_generic_extractor', False)
1331             self._downloader.report_warning(
1332                 '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
1333
1334         if not full_response:
1335             request = sanitized_Request(url)
1336             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
1337             # making it impossible to download only chunk of the file (yet we need only 512kB to
1338             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
1339             # that will always result in downloading the whole file that is not desirable.
1340             # Therefore for extraction pass we have to override Accept-Encoding to any in order
1341             # to accept raw bytes and being able to download only a chunk.
1342             # It may probably better to solve this by checking Content-Type for application/octet-stream
1343             # after HEAD request finishes, but not sure if we can rely on this.
1344             request.add_header('Accept-Encoding', '*')
1345             full_response = self._request_webpage(request, video_id)
1346
1347         first_bytes = full_response.read(512)
1348
1349         # Is it an M3U playlist?
1350         if first_bytes.startswith(b'#EXTM3U'):
1351             info_dict['formats'] = self._extract_m3u8_formats(url, video_id, 'mp4')
1352             self._sort_formats(info_dict['formats'])
1353             return info_dict
1354
1355         # Maybe it's a direct link to a video?
1356         # Be careful not to download the whole thing!
1357         if not is_html(first_bytes):
1358             self._downloader.report_warning(
1359                 'URL could be a direct video link, returning it as such.')
1360             info_dict.update({
1361                 'direct': True,
1362                 'url': url,
1363             })
1364             return info_dict
1365
1366         webpage = self._webpage_read_content(
1367             full_response, url, video_id, prefix=first_bytes)
1368
1369         self.report_extraction(video_id)
1370
1371         # Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest?
1372         try:
1373             doc = compat_etree_fromstring(webpage.encode('utf-8'))
1374             if doc.tag == 'rss':
1375                 return self._extract_rss(url, video_id, doc)
1376             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
1377                 smil = self._parse_smil(doc, url, video_id)
1378                 self._sort_formats(smil['formats'])
1379                 return smil
1380             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
1381                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
1382             elif re.match(r'(?i)^(?:{[^}]+})?MPD$', doc.tag):
1383                 info_dict['formats'] = self._parse_mpd_formats(
1384                     doc, video_id, mpd_base_url=url.rpartition('/')[0])
1385                 self._sort_formats(info_dict['formats'])
1386                 return info_dict
1387             elif re.match(r'^{http://ns\.adobe\.com/f4m/[12]\.0}manifest$', doc.tag):
1388                 info_dict['formats'] = self._parse_f4m_formats(doc, url, video_id)
1389                 self._sort_formats(info_dict['formats'])
1390                 return info_dict
1391         except compat_xml_parse_error:
1392             pass
1393
1394         # Is it a Camtasia project?
1395         camtasia_res = self._extract_camtasia(url, video_id, webpage)
1396         if camtasia_res is not None:
1397             return camtasia_res
1398
1399         # Sometimes embedded video player is hidden behind percent encoding
1400         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1401         # Unescaping the whole page allows to handle those cases in a generic way
1402         webpage = compat_urllib_parse_unquote(webpage)
1403
1404         # it's tempting to parse this further, but you would
1405         # have to take into account all the variations like
1406         #   Video Title - Site Name
1407         #   Site Name | Video Title
1408         #   Video Title - Tagline | Site Name
1409         # and so on and so forth; it's just not practical
1410         video_title = self._html_search_regex(
1411             r'(?s)<title>(.*?)</title>', webpage, 'video title',
1412             default='video')
1413
1414         # Try to detect age limit automatically
1415         age_limit = self._rta_search(webpage)
1416         # And then there are the jokers who advertise that they use RTA,
1417         # but actually don't.
1418         AGE_LIMIT_MARKERS = [
1419             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1420         ]
1421         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1422             age_limit = 18
1423
1424         # video uploader is domain name
1425         video_uploader = self._search_regex(
1426             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1427
1428         # Helper method
1429         def _playlist_from_matches(matches, getter=None, ie=None):
1430             urlrs = orderedSet(
1431                 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1432                 for m in matches)
1433             return self.playlist_result(
1434                 urlrs, playlist_id=video_id, playlist_title=video_title)
1435
1436         # Look for Brightcove Legacy Studio embeds
1437         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
1438         if bc_urls:
1439             self.to_screen('Brightcove video detected.')
1440             entries = [{
1441                 '_type': 'url',
1442                 'url': smuggle_url(bc_url, {'Referer': url}),
1443                 'ie_key': 'BrightcoveLegacy'
1444             } for bc_url in bc_urls]
1445
1446             return {
1447                 '_type': 'playlist',
1448                 'title': video_title,
1449                 'id': video_id,
1450                 'entries': entries,
1451             }
1452
1453         # Look for Brightcove New Studio embeds
1454         bc_urls = BrightcoveNewIE._extract_urls(webpage)
1455         if bc_urls:
1456             return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
1457
1458         # Look for embedded rtl.nl player
1459         matches = re.findall(
1460             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1461             webpage)
1462         if matches:
1463             return _playlist_from_matches(matches, ie='RtlNl')
1464
1465         vimeo_url = VimeoIE._extract_vimeo_url(url, webpage)
1466         if vimeo_url is not None:
1467             return self.url_result(vimeo_url)
1468
1469         vid_me_embed_url = self._search_regex(
1470             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
1471             webpage, 'vid.me embed', default=None)
1472         if vid_me_embed_url is not None:
1473             return self.url_result(vid_me_embed_url, 'Vidme')
1474
1475         # Look for embedded YouTube player
1476         matches = re.findall(r'''(?x)
1477             (?:
1478                 <iframe[^>]+?src=|
1479                 data-video-url=|
1480                 <embed[^>]+?src=|
1481                 embedSWF\(?:\s*|
1482                 new\s+SWFObject\(
1483             )
1484             (["\'])
1485                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1486                 (?:embed|v|p)/.+?)
1487             \1''', webpage)
1488         if matches:
1489             return _playlist_from_matches(
1490                 matches, lambda m: unescapeHTML(m[1]))
1491
1492         # Look for lazyYT YouTube embed
1493         matches = re.findall(
1494             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1495         if matches:
1496             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1497
1498         # Look for embedded Dailymotion player
1499         matches = re.findall(
1500             r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
1501         if matches:
1502             return _playlist_from_matches(
1503                 matches, lambda m: unescapeHTML(m[1]))
1504
1505         # Look for embedded Dailymotion playlist player (#3822)
1506         m = re.search(
1507             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1508         if m:
1509             playlists = re.findall(
1510                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1511             if playlists:
1512                 return _playlist_from_matches(
1513                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1514
1515         # Look for embedded Wistia player
1516         match = re.search(
1517             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1518         if match:
1519             embed_url = self._proto_relative_url(
1520                 unescapeHTML(match.group('url')))
1521             return {
1522                 '_type': 'url_transparent',
1523                 'url': embed_url,
1524                 'ie_key': 'Wistia',
1525                 'uploader': video_uploader,
1526                 'title': video_title,
1527                 'id': video_id,
1528             }
1529
1530         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1531         if match:
1532             return {
1533                 '_type': 'url_transparent',
1534                 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1535                 'ie_key': 'Wistia',
1536                 'uploader': video_uploader,
1537                 'title': video_title,
1538                 'id': match.group('id')
1539             }
1540
1541         # Look for SVT player
1542         svt_url = SVTIE._extract_url(webpage)
1543         if svt_url:
1544             return self.url_result(svt_url, 'SVT')
1545
1546         # Look for embedded condenast player
1547         matches = re.findall(
1548             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1549             webpage)
1550         if matches:
1551             return {
1552                 '_type': 'playlist',
1553                 'entries': [{
1554                     '_type': 'url',
1555                     'ie_key': 'CondeNast',
1556                     'url': ma,
1557                 } for ma in matches],
1558                 'title': video_title,
1559                 'id': video_id,
1560             }
1561
1562         # Look for Bandcamp pages with custom domain
1563         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1564         if mobj is not None:
1565             burl = unescapeHTML(mobj.group(1))
1566             # Don't set the extractor because it can be a track url or an album
1567             return self.url_result(burl)
1568
1569         # Look for embedded Vevo player
1570         mobj = re.search(
1571             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1572         if mobj is not None:
1573             return self.url_result(mobj.group('url'))
1574
1575         # Look for embedded Viddler player
1576         mobj = re.search(
1577             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1578             webpage)
1579         if mobj is not None:
1580             return self.url_result(mobj.group('url'))
1581
1582         # Look for NYTimes player
1583         mobj = re.search(
1584             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1585             webpage)
1586         if mobj is not None:
1587             return self.url_result(mobj.group('url'))
1588
1589         # Look for Libsyn player
1590         mobj = re.search(
1591             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1592         if mobj is not None:
1593             return self.url_result(mobj.group('url'))
1594
1595         # Look for Ooyala videos
1596         mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1597                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1598                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1599                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1600         if mobj is not None:
1601             return OoyalaIE._build_url_result(smuggle_url(mobj.group('ec'), {'domain': url}))
1602
1603         # Look for multiple Ooyala embeds on SBN network websites
1604         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1605         if mobj is not None:
1606             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1607             if embeds:
1608                 return _playlist_from_matches(
1609                     embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
1610
1611         # Look for Aparat videos
1612         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1613         if mobj is not None:
1614             return self.url_result(mobj.group(1), 'Aparat')
1615
1616         # Look for MPORA videos
1617         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1618         if mobj is not None:
1619             return self.url_result(mobj.group(1), 'Mpora')
1620
1621         # Look for embedded NovaMov-based player
1622         mobj = re.search(
1623             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1624                     (?P<url>http://(?:(?:embed|www)\.)?
1625                         (?:novamov\.com|
1626                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
1627                            videoweed\.(?:es|com)|
1628                            movshare\.(?:net|sx|ag)|
1629                            divxstage\.(?:eu|net|ch|co|at|ag))
1630                         /embed\.php.+?)\1''', webpage)
1631         if mobj is not None:
1632             return self.url_result(mobj.group('url'))
1633
1634         # Look for embedded Facebook player
1635         mobj = re.search(
1636             r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1637         if mobj is not None:
1638             return self.url_result(mobj.group('url'), 'Facebook')
1639
1640         # Look for embedded VK player
1641         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1642         if mobj is not None:
1643             return self.url_result(mobj.group('url'), 'VK')
1644
1645         # Look for embedded Odnoklassniki player
1646         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1', webpage)
1647         if mobj is not None:
1648             return self.url_result(mobj.group('url'), 'Odnoklassniki')
1649
1650         # Look for embedded ivi player
1651         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1652         if mobj is not None:
1653             return self.url_result(mobj.group('url'), 'Ivi')
1654
1655         # Look for embedded Huffington Post player
1656         mobj = re.search(
1657             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1658         if mobj is not None:
1659             return self.url_result(mobj.group('url'), 'HuffPost')
1660
1661         # Look for embed.ly
1662         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1663         if mobj is not None:
1664             return self.url_result(mobj.group('url'))
1665         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1666         if mobj is not None:
1667             return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
1668
1669         # Look for funnyordie embed
1670         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1671         if matches:
1672             return _playlist_from_matches(
1673                 matches, getter=unescapeHTML, ie='FunnyOrDie')
1674
1675         # Look for BBC iPlayer embed
1676         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1677         if matches:
1678             return _playlist_from_matches(matches, ie='BBCCoUk')
1679
1680         # Look for embedded RUTV player
1681         rutv_url = RUTVIE._extract_url(webpage)
1682         if rutv_url:
1683             return self.url_result(rutv_url, 'RUTV')
1684
1685         # Look for embedded TVC player
1686         tvc_url = TVCIE._extract_url(webpage)
1687         if tvc_url:
1688             return self.url_result(tvc_url, 'TVC')
1689
1690         # Look for embedded SportBox player
1691         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1692         if sportbox_urls:
1693             return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1694
1695         # Look for embedded PornHub player
1696         pornhub_url = PornHubIE._extract_url(webpage)
1697         if pornhub_url:
1698             return self.url_result(pornhub_url, 'PornHub')
1699
1700         # Look for embedded XHamster player
1701         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
1702         if xhamster_urls:
1703             return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
1704
1705         # Look for embedded TNAFlixNetwork player
1706         tnaflix_urls = TNAFlixNetworkEmbedIE._extract_urls(webpage)
1707         if tnaflix_urls:
1708             return _playlist_from_matches(tnaflix_urls, ie=TNAFlixNetworkEmbedIE.ie_key())
1709
1710         # Look for embedded Tvigle player
1711         mobj = re.search(
1712             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
1713         if mobj is not None:
1714             return self.url_result(mobj.group('url'), 'Tvigle')
1715
1716         # Look for embedded TED player
1717         mobj = re.search(
1718             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1719         if mobj is not None:
1720             return self.url_result(mobj.group('url'), 'TED')
1721
1722         # Look for embedded Ustream videos
1723         mobj = re.search(
1724             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1725         if mobj is not None:
1726             return self.url_result(mobj.group('url'), 'Ustream')
1727
1728         # Look for embedded arte.tv player
1729         mobj = re.search(
1730             r'<(?:script|iframe) [^>]*?src="(?P<url>http://www\.arte\.tv/(?:playerv2/embed|arte_vp/index)[^"]+)"',
1731             webpage)
1732         if mobj is not None:
1733             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1734
1735         # Look for embedded francetv player
1736         mobj = re.search(
1737             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
1738             webpage)
1739         if mobj is not None:
1740             return self.url_result(mobj.group('url'))
1741
1742         # Look for embedded smotri.com player
1743         smotri_url = SmotriIE._extract_url(webpage)
1744         if smotri_url:
1745             return self.url_result(smotri_url, 'Smotri')
1746
1747         # Look for embedded Myvi.ru player
1748         myvi_url = MyviIE._extract_url(webpage)
1749         if myvi_url:
1750             return self.url_result(myvi_url)
1751
1752         # Look for embedded soundcloud player
1753         mobj = re.search(
1754             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1755             webpage)
1756         if mobj is not None:
1757             url = unescapeHTML(mobj.group('url'))
1758             return self.url_result(url)
1759
1760         # Look for embedded vulture.com player
1761         mobj = re.search(
1762             r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1763             webpage)
1764         if mobj is not None:
1765             url = unescapeHTML(mobj.group('url'))
1766             return self.url_result(url, ie='Vulture')
1767
1768         # Look for embedded mtvservices player
1769         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
1770         if mtvservices_url:
1771             return self.url_result(mtvservices_url, ie='MTVServicesEmbedded')
1772
1773         # Look for embedded yahoo player
1774         mobj = re.search(
1775             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1776             webpage)
1777         if mobj is not None:
1778             return self.url_result(mobj.group('url'), 'Yahoo')
1779
1780         # Look for embedded sbs.com.au player
1781         mobj = re.search(
1782             r'''(?x)
1783             (?:
1784                 <meta\s+property="og:video"\s+content=|
1785                 <iframe[^>]+?src=
1786             )
1787             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1788             webpage)
1789         if mobj is not None:
1790             return self.url_result(mobj.group('url'), 'SBS')
1791
1792         # Look for embedded Cinchcast player
1793         mobj = re.search(
1794             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1795             webpage)
1796         if mobj is not None:
1797             return self.url_result(mobj.group('url'), 'Cinchcast')
1798
1799         mobj = re.search(
1800             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1801             webpage)
1802         if not mobj:
1803             mobj = re.search(
1804                 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
1805                 webpage)
1806         if mobj is not None:
1807             return self.url_result(mobj.group('url'), 'MLB')
1808
1809         mobj = re.search(
1810             r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1811             webpage)
1812         if mobj is not None:
1813             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1814
1815         mobj = re.search(
1816             r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1817             webpage)
1818         if mobj is not None:
1819             return self.url_result(mobj.group('url'), 'Livestream')
1820
1821         # Look for Zapiks embed
1822         mobj = re.search(
1823             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1824         if mobj is not None:
1825             return self.url_result(mobj.group('url'), 'Zapiks')
1826
1827         # Look for Kaltura embeds
1828         mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or
1829                 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))
1830         if mobj is not None:
1831             return self.url_result(smuggle_url(
1832                 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(),
1833                 {'source_url': url}), 'Kaltura')
1834
1835         # Look for Eagle.Platform embeds
1836         mobj = re.search(
1837             r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1838         if mobj is not None:
1839             return self.url_result(mobj.group('url'), 'EaglePlatform')
1840
1841         # Look for ClipYou (uses Eagle.Platform) embeds
1842         mobj = re.search(
1843             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1844         if mobj is not None:
1845             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1846
1847         # Look for Pladform embeds
1848         pladform_url = PladformIE._extract_url(webpage)
1849         if pladform_url:
1850             return self.url_result(pladform_url)
1851
1852         # Look for Videomore embeds
1853         videomore_url = VideomoreIE._extract_url(webpage)
1854         if videomore_url:
1855             return self.url_result(videomore_url)
1856
1857         # Look for Playwire embeds
1858         mobj = re.search(
1859             r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
1860         if mobj is not None:
1861             return self.url_result(mobj.group('url'))
1862
1863         # Look for 5min embeds
1864         mobj = re.search(
1865             r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1866         if mobj is not None:
1867             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1868
1869         # Look for Crooks and Liars embeds
1870         mobj = re.search(
1871             r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1872         if mobj is not None:
1873             return self.url_result(mobj.group('url'))
1874
1875         # Look for NBC Sports VPlayer embeds
1876         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1877         if nbc_sports_url:
1878             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1879
1880         # Look for Google Drive embeds
1881         google_drive_url = GoogleDriveIE._extract_url(webpage)
1882         if google_drive_url:
1883             return self.url_result(google_drive_url, 'GoogleDrive')
1884
1885         # Look for UDN embeds
1886         mobj = re.search(
1887             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage)
1888         if mobj is not None:
1889             return self.url_result(
1890                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1891
1892         # Look for Senate ISVP iframe
1893         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
1894         if senate_isvp_url:
1895             return self.url_result(senate_isvp_url, 'SenateISVP')
1896
1897         # Look for Dailymotion Cloud videos
1898         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
1899         if dmcloud_url:
1900             return self.url_result(dmcloud_url, 'DailymotionCloud')
1901
1902         # Look for OnionStudios embeds
1903         onionstudios_url = OnionStudiosIE._extract_url(webpage)
1904         if onionstudios_url:
1905             return self.url_result(onionstudios_url)
1906
1907         # Look for SnagFilms embeds
1908         snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage)
1909         if snagfilms_url:
1910             return self.url_result(snagfilms_url)
1911
1912         # Look for JWPlatform embeds
1913         jwplatform_url = JWPlatformIE._extract_url(webpage)
1914         if jwplatform_url:
1915             return self.url_result(jwplatform_url, 'JWPlatform')
1916
1917         # Look for ScreenwaveMedia embeds
1918         mobj = re.search(ScreenwaveMediaIE.EMBED_PATTERN, webpage)
1919         if mobj is not None:
1920             return self.url_result(unescapeHTML(mobj.group('url')), 'ScreenwaveMedia')
1921
1922         # Look for Digiteka embeds
1923         digiteka_url = DigitekaIE._extract_url(webpage)
1924         if digiteka_url:
1925             return self.url_result(self._proto_relative_url(digiteka_url), DigitekaIE.ie_key())
1926
1927         # Look for Limelight embeds
1928         mobj = re.search(r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})', webpage)
1929         if mobj:
1930             lm = {
1931                 'Media': 'media',
1932                 'Channel': 'channel',
1933                 'ChannelList': 'channel_list',
1934             }
1935             return self.url_result('limelight:%s:%s' % (
1936                 lm[mobj.group(1)], mobj.group(2)), 'Limelight%s' % mobj.group(1), mobj.group(2))
1937
1938         # Look for AdobeTVVideo embeds
1939         mobj = re.search(
1940             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
1941             webpage)
1942         if mobj is not None:
1943             return self.url_result(
1944                 self._proto_relative_url(unescapeHTML(mobj.group(1))),
1945                 'AdobeTVVideo')
1946
1947         # Look for Vine embeds
1948         mobj = re.search(
1949             r'<iframe[^>]+src=[\'"]((?:https?:)?//(?:www\.)?vine\.co/v/[^/]+/embed/(?:simple|postcard))',
1950             webpage)
1951         if mobj is not None:
1952             return self.url_result(
1953                 self._proto_relative_url(unescapeHTML(mobj.group(1))), 'Vine')
1954
1955         # Look for Instagram embeds
1956         instagram_embed_url = InstagramIE._extract_embed_url(webpage)
1957         if instagram_embed_url is not None:
1958             return self.url_result(instagram_embed_url, InstagramIE.ie_key())
1959
1960         # Look for LiveLeak embeds
1961         liveleak_url = LiveLeakIE._extract_url(webpage)
1962         if liveleak_url:
1963             return self.url_result(liveleak_url, 'LiveLeak')
1964
1965         def check_video(vurl):
1966             if YoutubeIE.suitable(vurl):
1967                 return True
1968             vpath = compat_urlparse.urlparse(vurl).path
1969             vext = determine_ext(vpath)
1970             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1971
1972         def filter_video(urls):
1973             return list(filter(check_video, urls))
1974
1975         # Start with something easy: JW Player in SWFObject
1976         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1977         if not found:
1978             # Look for gorilla-vid style embedding
1979             found = filter_video(re.findall(r'''(?sx)
1980                 (?:
1981                     jw_plugins|
1982                     JWPlayerOptions|
1983                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1984                 )
1985                 .*?
1986                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1987         if not found:
1988             # Broaden the search a little bit
1989             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1990         if not found:
1991             # Broaden the findall a little bit: JWPlayer JS loader
1992             found = filter_video(re.findall(
1993                 r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1994         if not found:
1995             # Flow player
1996             found = filter_video(re.findall(r'''(?xs)
1997                 flowplayer\("[^"]+",\s*
1998                     \{[^}]+?\}\s*,
1999                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
2000                         ["']?url["']?\s*:\s*["']([^"']+)["']
2001             ''', webpage))
2002         if not found:
2003             # Cinerama player
2004             found = re.findall(
2005                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
2006         if not found:
2007             # Try to find twitter cards info
2008             found = filter_video(re.findall(
2009                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
2010         if not found:
2011             # We look for Open Graph info:
2012             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
2013             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
2014             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
2015             if m_video_type is not None:
2016                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
2017         if not found:
2018             # HTML5 video
2019             found = re.findall(r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
2020         if not found:
2021             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
2022             found = re.search(
2023                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
2024                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
2025                 webpage)
2026             if not found:
2027                 # Look also in Refresh HTTP header
2028                 refresh_header = head_response.headers.get('Refresh')
2029                 if refresh_header:
2030                     # In python 2 response HTTP headers are bytestrings
2031                     if sys.version_info < (3, 0) and isinstance(refresh_header, str):
2032                         refresh_header = refresh_header.decode('iso-8859-1')
2033                     found = re.search(REDIRECT_REGEX, refresh_header)
2034             if found:
2035                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
2036                 self.report_following_redirect(new_url)
2037                 return {
2038                     '_type': 'url',
2039                     'url': new_url,
2040                 }
2041         if not found:
2042             raise UnsupportedError(url)
2043
2044         entries = []
2045         for video_url in found:
2046             video_url = video_url.replace('\\/', '/')
2047             video_url = compat_urlparse.urljoin(url, video_url)
2048             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
2049
2050             # Sometimes, jwplayer extraction will result in a YouTube URL
2051             if YoutubeIE.suitable(video_url):
2052                 entries.append(self.url_result(video_url, 'Youtube'))
2053                 continue
2054
2055             # here's a fun little line of code for you:
2056             video_id = os.path.splitext(video_id)[0]
2057
2058             entry_info_dict = {
2059                 'id': video_id,
2060                 'uploader': video_uploader,
2061                 'title': video_title,
2062                 'age_limit': age_limit,
2063             }
2064
2065             ext = determine_ext(video_url)
2066             if ext == 'smil':
2067                 entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
2068             elif ext == 'xspf':
2069                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
2070             elif ext == 'm3u8':
2071                 entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
2072             elif ext == 'mpd':
2073                 entry_info_dict['formats'] = self._extract_mpd_formats(video_url, video_id)
2074             elif ext == 'f4m':
2075                 entry_info_dict['formats'] = self._extract_f4m_formats(video_url, video_id)
2076             else:
2077                 entry_info_dict['url'] = video_url
2078
2079             if entry_info_dict.get('formats'):
2080                 self._sort_formats(entry_info_dict['formats'])
2081
2082             entries.append(entry_info_dict)
2083
2084         if len(entries) == 1:
2085             return entries[0]
2086         else:
2087             for num, e in enumerate(entries, start=1):
2088                 # 'url' results don't have a title
2089                 if e.get('title') is not None:
2090                     e['title'] = '%s (%d)' % (e['title'], num)
2091             return {
2092                 '_type': 'playlist',
2093                 'entries': entries,
2094             }