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

   1 # coding: 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     xpath_text,
  31 )
  32 from .brightcove import (
  33     BrightcoveLegacyIE,
  34     BrightcoveNewIE,
  35 )
  36 from .nbc import NBCSportsVPlayerIE
  37 from .ooyala import OoyalaIE
  38 from .rutv import RUTVIE
  39 from .tvc import TVCIE
  40 from .sportbox import SportBoxEmbedIE
  41 from .smotri import SmotriIE
  42 from .myvi import MyviIE
  43 from .condenast import CondeNastIE
  44 from .udn import UDNEmbedIE
  45 from .senateisvp import SenateISVPIE
  46 from .svt import SVTIE
  47 from .pornhub import PornHubIE
  48 from .xhamster import XHamsterEmbedIE
  49 from .tnaflix import TNAFlixNetworkEmbedIE
  50 from .vimeo import VimeoIE
  51 from .dailymotion import (
  52     DailymotionIE,
  53     DailymotionCloudIE,
  54 )
  55 from .onionstudios import OnionStudiosIE
  56 from .viewlift import ViewLiftEmbedIE
  57 from .screenwavemedia import ScreenwaveMediaIE
  58 from .mtv import MTVServicesEmbeddedIE
  59 from .pladform import PladformIE
  60 from .videomore import VideomoreIE
  61 from .googledrive import GoogleDriveIE
  62 from .jwplatform import JWPlatformIE
  63 from .digiteka import DigitekaIE
  64 from .arkena import ArkenaIE
  65 from .instagram import InstagramIE
  66 from .liveleak import LiveLeakIE
  67 from .threeqsdn import ThreeQSDNIE
  68 from .theplatform import ThePlatformIE
  69 from .vessel import VesselIE
  70 from .kaltura import KalturaIE
  71 from .eagleplatform import EaglePlatformIE
  72 from .facebook import FacebookIE
  73 from .soundcloud import SoundcloudIE
  74 from .vbox7 import Vbox7IE
  75 from .dbtv import DBTVIE
  76
  77
  78 class GenericIE(InfoExtractor):
  79     IE_DESC = 'Generic downloader that works on some sites'
  80     _VALID_URL = r'.*'
  81     IE_NAME = 'generic'
  82     _TESTS = [
  83         # Direct link to a video
  84         {
  85             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  86             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  87             'info_dict': {
  88                 'id': 'trailer',
  89                 'ext': 'mp4',
  90                 'title': 'trailer',
  91                 'upload_date': '20100513',
  92             }
  93         },
  94         # Direct link to media delivered compressed (until Accept-Encoding is *)
  95         {
  96             'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
  97             'md5': '128c42e68b13950268b648275386fc74',
  98             'info_dict': {
  99                 'id': 'FictionJunction-Parallel_Hearts',
 100                 'ext': 'flac',
 101                 'title': 'FictionJunction-Parallel_Hearts',
 102                 'upload_date': '20140522',
 103             },
 104             'expected_warnings': [
 105                 'URL could be a direct video link, returning it as such.'
 106             ],
 107             'skip': 'URL invalid',
 108         },
 109         # Direct download with broken HEAD
 110         {
 111             'url': 'http://ai-radio.org:8000/radio.opus',
 112             'info_dict': {
 113                 'id': 'radio',
 114                 'ext': 'opus',
 115                 'title': 'radio',
 116             },
 117             'params': {
 118                 'skip_download': True,  # infinite live stream
 119             },
 120             'expected_warnings': [
 121                 r'501.*Not Implemented',
 122                 r'400.*Bad Request',
 123             ],
 124         },
 125         # Direct link with incorrect MIME type
 126         {
 127             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 128             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 129             'info_dict': {
 130                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 131                 'id': '5_Lennart_Poettering_-_Systemd',
 132                 'ext': 'webm',
 133                 'title': '5_Lennart_Poettering_-_Systemd',
 134                 'upload_date': '20141120',
 135             },
 136             'expected_warnings': [
 137                 'URL could be a direct video link, returning it as such.'
 138             ]
 139         },
 140         # RSS feed
 141         {
 142             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 143             'info_dict': {
 144                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 145                 'title': 'Zero Punctuation',
 146                 'description': 're:.*groundbreaking video review series.*'
 147             },
 148             'playlist_mincount': 11,
 149         },
 150         # RSS feed with enclosure
 151         {
 152             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 153             'info_dict': {
 154                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 155                 'ext': 'm4v',
 156                 'upload_date': '20150228',
 157                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 158             }
 159         },
 160         # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
 161         {
 162             'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
 163             'info_dict': {
 164                 'id': 'smil',
 165                 'ext': 'mp4',
 166                 'title': 'Automatics, robotics and biocybernetics',
 167                 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
 168                 'upload_date': '20130627',
 169                 'formats': 'mincount:16',
 170                 'subtitles': 'mincount:1',
 171             },
 172             'params': {
 173                 'force_generic_extractor': True,
 174                 'skip_download': True,
 175             },
 176         },
 177         # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html
 178         {
 179             'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil',
 180             'info_dict': {
 181                 'id': 'hds',
 182                 'ext': 'flv',
 183                 'title': 'hds',
 184                 'formats': 'mincount:1',
 185             },
 186             'params': {
 187                 'skip_download': True,
 188             },
 189         },
 190         # SMIL from https://www.restudy.dk/video/play/id/1637
 191         {
 192             'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml',
 193             'info_dict': {
 194                 'id': 'video_1637',
 195                 'ext': 'flv',
 196                 'title': 'video_1637',
 197                 'formats': 'mincount:3',
 198             },
 199             'params': {
 200                 'skip_download': True,
 201             },
 202         },
 203         # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm
 204         {
 205             'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil',
 206             'info_dict': {
 207                 'id': 'smil-service',
 208                 'ext': 'flv',
 209                 'title': 'smil-service',
 210                 'formats': 'mincount:1',
 211             },
 212             'params': {
 213                 'skip_download': True,
 214             },
 215         },
 216         # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370
 217         {
 218             'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil',
 219             'info_dict': {
 220                 'id': '4719370',
 221                 'ext': 'mp4',
 222                 'title': '571de1fd-47bc-48db-abf9-238872a58d1f',
 223                 'formats': 'mincount:3',
 224             },
 225             'params': {
 226                 'skip_download': True,
 227             },
 228         },
 229         # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html
 230         {
 231             'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf',
 232             'info_dict': {
 233                 'id': 'mZlp2ctYIUEB',
 234                 'ext': 'mp4',
 235                 'title': 'Tikibad ontruimd wegens brand',
 236                 'description': 'md5:05ca046ff47b931f9b04855015e163a4',
 237                 'thumbnail': 're:^https?://.*\.jpg$',
 238                 'duration': 33,
 239             },
 240             'params': {
 241                 'skip_download': True,
 242             },
 243         },
 244         # MPD from http://dash-mse-test.appspot.com/media.html
 245         {
 246             'url': 'http://yt-dash-mse-test.commondatastorage.googleapis.com/media/car-20120827-manifest.mpd',
 247             'md5': '4b57baab2e30d6eb3a6a09f0ba57ef53',
 248             'info_dict': {
 249                 'id': 'car-20120827-manifest',
 250                 'ext': 'mp4',
 251                 'title': 'car-20120827-manifest',
 252                 'formats': 'mincount:9',
 253                 'upload_date': '20130904',
 254             },
 255             'params': {
 256                 'format': 'bestvideo',
 257             },
 258         },
 259         # m3u8 served with Content-Type: audio/x-mpegURL; charset=utf-8
 260         {
 261             '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',
 262             'info_dict': {
 263                 'id': 'content',
 264                 'ext': 'mp4',
 265                 'title': 'content',
 266                 'formats': 'mincount:8',
 267             },
 268             'params': {
 269                 # m3u8 downloads
 270                 'skip_download': True,
 271             },
 272             'skip': 'video gone',
 273         },
 274         # m3u8 served with Content-Type: text/plain
 275         {
 276             'url': 'http://www.nacentapps.com/m3u8/index.m3u8',
 277             'info_dict': {
 278                 'id': 'index',
 279                 'ext': 'mp4',
 280                 'title': 'index',
 281                 'upload_date': '20140720',
 282                 'formats': 'mincount:11',
 283             },
 284             'params': {
 285                 # m3u8 downloads
 286                 'skip_download': True,
 287             },
 288             'skip': 'video gone',
 289         },
 290         # google redirect
 291         {
 292             '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',
 293             'info_dict': {
 294                 'id': 'cmQHVoWB5FY',
 295                 'ext': 'mp4',
 296                 'upload_date': '20130224',
 297                 'uploader_id': 'TheVerge',
 298                 'description': 're:^Chris Ziegler takes a look at the\.*',
 299                 'uploader': 'The Verge',
 300                 'title': 'First Firefox OS phones side-by-side',
 301             },
 302             'params': {
 303                 'skip_download': False,
 304             }
 305         },
 306         {
 307             # redirect in Refresh HTTP header
 308             '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',
 309             'info_dict': {
 310                 'id': 'pO8h3EaFRdo',
 311                 'ext': 'mp4',
 312                 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
 313                 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5',
 314                 'upload_date': '20150917',
 315                 'uploader_id': 'brtvofficial',
 316                 'uploader': 'Boiler Room',
 317             },
 318             'params': {
 319                 'skip_download': False,
 320             },
 321         },
 322         {
 323             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
 324             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
 325             'info_dict': {
 326                 'id': '13601338388002',
 327                 'ext': 'mp4',
 328                 'uploader': 'www.hodiho.fr',
 329                 'title': 'R\u00e9gis plante sa Jeep',
 330             }
 331         },
 332         # bandcamp page with custom domain
 333         {
 334             'add_ie': ['Bandcamp'],
 335             'url': 'http://bronyrock.com/track/the-pony-mash',
 336             'info_dict': {
 337                 'id': '3235767654',
 338                 'ext': 'mp3',
 339                 'title': 'The Pony Mash',
 340                 'uploader': 'M_Pallante',
 341             },
 342             'skip': 'There is a limit of 200 free downloads / month for the test song',
 343         },
 344         # embedded brightcove video
 345         # it also tests brightcove videos that need to set the 'Referer' in the
 346         # http requests
 347         {
 348             'add_ie': ['BrightcoveLegacy'],
 349             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
 350             'info_dict': {
 351                 'id': '2765128793001',
 352                 'ext': 'mp4',
 353                 'title': 'Le cours de bourse : l’analyse technique',
 354                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
 355                 'uploader': 'BFM BUSINESS',
 356             },
 357             'params': {
 358                 'skip_download': True,
 359             },
 360         },
 361         {
 362             # https://github.com/rg3/youtube-dl/issues/2253
 363             'url': 'http://bcove.me/i6nfkrc3',
 364             'md5': '0ba9446db037002366bab3b3eb30c88c',
 365             'info_dict': {
 366                 'id': '3101154703001',
 367                 'ext': 'mp4',
 368                 'title': 'Still no power',
 369                 'uploader': 'thestar.com',
 370                 '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.',
 371             },
 372             'add_ie': ['BrightcoveLegacy'],
 373             'skip': 'video gone',
 374         },
 375         {
 376             'url': 'http://www.championat.com/video/football/v/87/87499.html',
 377             'md5': 'fb973ecf6e4a78a67453647444222983',
 378             'info_dict': {
 379                 'id': '3414141473001',
 380                 'ext': 'mp4',
 381                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 382                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 383                 'uploader': 'Championat',
 384             },
 385         },
 386         {
 387             # https://github.com/rg3/youtube-dl/issues/3541
 388             'add_ie': ['BrightcoveLegacy'],
 389             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 390             'info_dict': {
 391                 'id': '3866516442001',
 392                 'ext': 'mp4',
 393                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 394                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 395                 'uploader': 'SBS Broadcasting',
 396             },
 397             'skip': 'Restricted to Netherlands',
 398             'params': {
 399                 'skip_download': True,  # m3u8 download
 400             },
 401         },
 402         # ooyala video
 403         {
 404             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 405             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 406             'info_dict': {
 407                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 408                 'ext': 'mp4',
 409                 'title': '2cc213299525360.mov',  # that's what we get
 410                 'duration': 238.231,
 411             },
 412             'add_ie': ['Ooyala'],
 413         },
 414         {
 415             # ooyala video embedded with http://player.ooyala.com/iframe.js
 416             'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/',
 417             'info_dict': {
 418                 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB',
 419                 'ext': 'mp4',
 420                 'title': '"Steve Jobs: Man in the Machine" trailer',
 421                 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
 422                 'duration': 135.427,
 423             },
 424             'params': {
 425                 'skip_download': True,
 426             },
 427             'skip': 'movie expired',
 428         },
 429         # embed.ly video
 430         {
 431             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 432             'info_dict': {
 433                 'id': '9ODmcdjQcHQ',
 434                 'ext': 'mp4',
 435                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 436                 'upload_date': '20140225',
 437                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 438                 'uploader': 'Tested',
 439                 'uploader_id': 'testedcom',
 440             },
 441             # No need to test YoutubeIE here
 442             'params': {
 443                 'skip_download': True,
 444             },
 445         },
 446         # funnyordie embed
 447         {
 448             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 449             'info_dict': {
 450                 'id': '18e820ec3f',
 451                 'ext': 'mp4',
 452                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 453                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 454             },
 455             # HEAD requests lead to endless 301, while GET is OK
 456             'expected_warnings': ['301'],
 457         },
 458         # RUTV embed
 459         {
 460             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 461             'info_dict': {
 462                 'id': '776940',
 463                 'ext': 'mp4',
 464                 'title': 'Охотское море стало целиком российским',
 465                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 466             },
 467             'params': {
 468                 # m3u8 download
 469                 'skip_download': True,
 470             },
 471         },
 472         # TVC embed
 473         {
 474             '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/',
 475             'info_dict': {
 476                 'id': '55304',
 477                 'ext': 'mp4',
 478                 'title': 'Дошкольное воспитание',
 479             },
 480         },
 481         # SportBox embed
 482         {
 483             'url': 'http://www.vestifinance.ru/articles/25753',
 484             'info_dict': {
 485                 'id': '25753',
 486                 'title': 'Прямые трансляции с Форума-выставки "Госзаказ-2013"',
 487             },
 488             'playlist': [{
 489                 'info_dict': {
 490                     'id': '370908',
 491                     'title': 'Госзаказ. День 3',
 492                     'ext': 'mp4',
 493                 }
 494             }, {
 495                 'info_dict': {
 496                     'id': '370905',
 497                     'title': 'Госзаказ. День 2',
 498                     'ext': 'mp4',
 499                 }
 500             }, {
 501                 'info_dict': {
 502                     'id': '370902',
 503                     'title': 'Госзаказ. День 1',
 504                     'ext': 'mp4',
 505                 }
 506             }],
 507             'params': {
 508                 # m3u8 download
 509                 'skip_download': True,
 510             },
 511         },
 512         # Myvi.ru embed
 513         {
 514             'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
 515             'info_dict': {
 516                 'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
 517                 'ext': 'mp4',
 518                 'title': 'Ужастики, русский трейлер (2015)',
 519                 'thumbnail': 're:^https?://.*\.jpg$',
 520                 'duration': 153,
 521             }
 522         },
 523         # XHamster embed
 524         {
 525             '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',
 526             'info_dict': {
 527                 'id': 'showthread',
 528                 'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
 529             },
 530             'playlist_mincount': 7,
 531             # This forum does not allow <iframe> syntaxes anymore
 532             # Now HTML tags are displayed as-is
 533             'skip': 'No videos on this page',
 534         },
 535         # Embedded TED video
 536         {
 537             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 538             'md5': '65fdff94098e4a607385a60c5177c638',
 539             'info_dict': {
 540                 'id': '1969',
 541                 'ext': 'mp4',
 542                 'title': 'Hidden miracles of the natural world',
 543                 'uploader': 'Louie Schwartzberg',
 544                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 545             }
 546         },
 547         # Embedded Ustream video
 548         {
 549             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 550             'md5': '27b99cdb639c9b12a79bca876a073417',
 551             'info_dict': {
 552                 'id': '45734260',
 553                 'ext': 'flv',
 554                 'uploader': 'AU SPA:  The NSA and Privacy',
 555                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 556             }
 557         },
 558         # nowvideo embed hidden behind percent encoding
 559         {
 560             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 561             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 562             'info_dict': {
 563                 'id': '06e53103ca9aa',
 564                 'ext': 'flv',
 565                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 566                 'description': 'No description',
 567             },
 568         },
 569         # arte embed
 570         {
 571             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 572             'md5': '7653032cbb25bf6c80d80f217055fa43',
 573             'info_dict': {
 574                 'id': '048195-004_PLUS7-F',
 575                 'ext': 'flv',
 576                 'title': 'X:enius',
 577                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 578                 'upload_date': '20140320',
 579             },
 580             'params': {
 581                 'skip_download': 'Requires rtmpdump'
 582             },
 583             'skip': 'video gone',
 584         },
 585         # francetv embed
 586         {
 587             'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
 588             'info_dict': {
 589                 'id': 'EV_30231',
 590                 'ext': 'mp4',
 591                 'title': 'Alcaline, le concert avec Calogero',
 592                 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
 593                 'upload_date': '20150226',
 594                 'timestamp': 1424989860,
 595                 'duration': 5400,
 596             },
 597             'params': {
 598                 # m3u8 downloads
 599                 'skip_download': True,
 600             },
 601             'expected_warnings': [
 602                 'Forbidden'
 603             ]
 604         },
 605         # Condé Nast embed
 606         {
 607             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 608             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 609             'info_dict': {
 610                 'id': '53501be369702d3275860000',
 611                 'ext': 'mp4',
 612                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 613             }
 614         },
 615         # Dailymotion embed
 616         {
 617             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 618             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 619             'info_dict': {
 620                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 621                 'ext': 'mp4',
 622                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 623                 'description': 'md5:faf028e48a461b8b7fad38f1e104b119',
 624                 'uploader': 'Spi0n',
 625                 'uploader_id': 'xgditw',
 626                 'upload_date': '20140425',
 627                 'timestamp': 1398441542,
 628             },
 629             'add_ie': ['Dailymotion'],
 630         },
 631         # YouTube embed
 632         {
 633             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 634             'info_dict': {
 635                 'id': 'FXRb4ykk4S0',
 636                 'ext': 'mp4',
 637                 'title': 'The NBL Auction 2014',
 638                 'uploader': 'BADMINTON England',
 639                 'uploader_id': 'BADMINTONEvents',
 640                 'upload_date': '20140603',
 641                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 642             },
 643             'add_ie': ['Youtube'],
 644             'params': {
 645                 'skip_download': True,
 646             }
 647         },
 648         # MTVSercices embed
 649         {
 650             'url': 'http://www.vulture.com/2016/06/new-key-peele-sketches-released.html',
 651             'md5': 'ca1aef97695ef2c1d6973256a57e5252',
 652             'info_dict': {
 653                 'id': '769f7ec0-0692-4d62-9b45-0d88074bffc1',
 654                 'ext': 'mp4',
 655                 'title': 'Key and Peele|October 10, 2012|2|203|Liam Neesons - Uncensored',
 656                 'description': 'Two valets share their love for movie star Liam Neesons.',
 657                 'timestamp': 1349922600,
 658                 'upload_date': '20121011',
 659             },
 660         },
 661         # YouTube embed via <data-embed-url="">
 662         {
 663             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 664             'info_dict': {
 665                 'id': '4vAffPZIT44',
 666                 'ext': 'mp4',
 667                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 668                 'uploader': 'Gameloft',
 669                 'uploader_id': 'gameloft',
 670                 'upload_date': '20140828',
 671                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 672             },
 673             'params': {
 674                 'skip_download': True,
 675             }
 676         },
 677         # Camtasia studio
 678         {
 679             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 680             'playlist': [{
 681                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 682                 'info_dict': {
 683                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 684                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 685                     'ext': 'flv',
 686                     'duration': 2235.90,
 687                 }
 688             }, {
 689                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 690                 'info_dict': {
 691                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 692                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 693                     'ext': 'flv',
 694                     'duration': 2235.93,
 695                 }
 696             }],
 697             'info_dict': {
 698                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 699             }
 700         },
 701         # Flowplayer
 702         {
 703             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 704             'md5': '9d65602bf31c6e20014319c7d07fba27',
 705             'info_dict': {
 706                 'id': '5123ea6d5e5a7',
 707                 'ext': 'mp4',
 708                 'age_limit': 18,
 709                 'uploader': 'www.handjobhub.com',
 710                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 711             }
 712         },
 713         # Multiple brightcove videos
 714         # https://github.com/rg3/youtube-dl/issues/2283
 715         {
 716             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 717             'info_dict': {
 718                 'id': 'always-never',
 719                 'title': 'Always / Never - The New Yorker',
 720             },
 721             'playlist_count': 3,
 722             'params': {
 723                 'extract_flat': False,
 724                 'skip_download': True,
 725             }
 726         },
 727         # MLB embed
 728         {
 729             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 730             'md5': '96f09a37e44da40dd083e12d9a683327',
 731             'info_dict': {
 732                 'id': '33322633',
 733                 'ext': 'mp4',
 734                 'title': 'Ump changes call to ball',
 735                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 736                 'duration': 48,
 737                 'timestamp': 1401537900,
 738                 'upload_date': '20140531',
 739                 'thumbnail': 're:^https?://.*\.jpg$',
 740             },
 741         },
 742         # Wistia embed
 743         {
 744             'url': 'http://study.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 745             'md5': '1953f3a698ab51cfc948ed3992a0b7ff',
 746             'info_dict': {
 747                 'id': '6e2wtrbdaf',
 748                 'ext': 'mov',
 749                 'title': 'paywall_north-american-exploration-failed-colonies-of-spain-france-england',
 750                 'description': 'a Paywall Videos video from Remilon',
 751                 'duration': 644.072,
 752                 'uploader': 'study.com',
 753                 'timestamp': 1459678540,
 754                 'upload_date': '20160403',
 755                 'filesize': 24687186,
 756             },
 757         },
 758         {
 759             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 760             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 761             'info_dict': {
 762                 'id': 'uxjb0lwrcz',
 763                 'ext': 'mp4',
 764                 'title': 'Conversation about Hexagonal Rails Part 1',
 765                 'description': 'a Martin Fowler video from ThoughtWorks',
 766                 'duration': 1715.0,
 767                 'uploader': 'thoughtworks.wistia.com',
 768                 'timestamp': 1401832161,
 769                 'upload_date': '20140603',
 770             },
 771         },
 772         # Wistia standard embed (async)
 773         {
 774             'url': 'https://www.getdrip.com/university/brennan-dunn-drip-workshop/',
 775             'info_dict': {
 776                 'id': '807fafadvk',
 777                 'ext': 'mp4',
 778                 'title': 'Drip Brennan Dunn Workshop',
 779                 'description': 'a JV Webinars video from getdrip-1',
 780                 'duration': 4986.95,
 781                 'timestamp': 1463607249,
 782                 'upload_date': '20160518',
 783             },
 784             'params': {
 785                 'skip_download': True,
 786             }
 787         },
 788         # Soundcloud embed
 789         {
 790             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 791             'info_dict': {
 792                 'id': '174391317',
 793                 'ext': 'mp3',
 794                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 795                 'uploader': 'Sophos Security',
 796                 'title': 'Chet Chat 171 - Oct 29, 2014',
 797                 'upload_date': '20141029',
 798             }
 799         },
 800         # Soundcloud multiple embeds
 801         {
 802             'url': 'http://www.guitarplayer.com/lessons/1014/legato-workout-one-hour-to-more-fluid-performance---tab/52809',
 803             'info_dict': {
 804                 'id': '52809',
 805                 'title': 'Guitar Essentials: Legato Workout—One-Hour to Fluid Performance  | TAB + AUDIO',
 806             },
 807             'playlist_mincount': 7,
 808         },
 809         # Livestream embed
 810         {
 811             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 812             'info_dict': {
 813                 'id': '67864563',
 814                 'ext': 'flv',
 815                 'upload_date': '20141112',
 816                 'title': 'Rosetta #CometLanding webcast HL 10',
 817             }
 818         },
 819         # Another Livestream embed, without 'new.' in URL
 820         {
 821             'url': 'https://www.freespeech.org/',
 822             'info_dict': {
 823                 'id': '123537347',
 824                 'ext': 'mp4',
 825                 'title': 're:^FSTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 826             },
 827             'params': {
 828                 # Live stream
 829                 'skip_download': True,
 830             },
 831         },
 832         # LazyYT
 833         {
 834             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
 835             'info_dict': {
 836                 'id': '1986',
 837                 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
 838             },
 839             'playlist_mincount': 2,
 840         },
 841         # Cinchcast embed
 842         {
 843             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
 844             'info_dict': {
 845                 'id': '7141703',
 846                 'ext': 'mp3',
 847                 'upload_date': '20141126',
 848                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
 849             }
 850         },
 851         # Cinerama player
 852         {
 853             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
 854             'info_dict': {
 855                 'id': '730m_DandD_1901_512k',
 856                 'ext': 'mp4',
 857                 'uploader': 'www.abc.net.au',
 858                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
 859             }
 860         },
 861         # embedded viddler video
 862         {
 863             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
 864             'info_dict': {
 865                 'id': '4d03aad9',
 866                 'ext': 'mp4',
 867                 'uploader': 'deadspin',
 868                 'title': 'WALL-TO-GORTAT',
 869                 'timestamp': 1422285291,
 870                 'upload_date': '20150126',
 871             },
 872             'add_ie': ['Viddler'],
 873         },
 874         # Libsyn embed
 875         {
 876             'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
 877             'info_dict': {
 878                 'id': '3377616',
 879                 'ext': 'mp3',
 880                 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
 881                 'description': 'md5:601cb790edd05908957dae8aaa866465',
 882                 'upload_date': '20150220',
 883             },
 884             'skip': 'All The Daily Show URLs now redirect to http://www.cc.com/shows/',
 885         },
 886         # jwplayer YouTube
 887         {
 888             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
 889             'info_dict': {
 890                 'id': 'Mrj4DVp2zeA',
 891                 'ext': 'mp4',
 892                 'upload_date': '20150212',
 893                 'uploader': 'The National Archives UK',
 894                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
 895                 'uploader_id': 'NationalArchives08',
 896                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
 897             },
 898         },
 899         # rtl.nl embed
 900         {
 901             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
 902             'playlist_mincount': 5,
 903             'info_dict': {
 904                 'id': 'aanslagen-kopenhagen',
 905                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
 906             }
 907         },
 908         # Zapiks embed
 909         {
 910             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
 911             'info_dict': {
 912                 'id': '118046',
 913                 'ext': 'mp4',
 914                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
 915             }
 916         },
 917         # Kaltura embed (different embed code)
 918         {
 919             'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
 920             'info_dict': {
 921                 'id': '1_a52wc67y',
 922                 'ext': 'flv',
 923                 'upload_date': '20150127',
 924                 'uploader_id': 'PremierMedia',
 925                 'timestamp': int,
 926                 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
 927             },
 928         },
 929         # Kaltura embed protected with referrer
 930         {
 931             'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero',
 932             'info_dict': {
 933                 'id': '1_g4fbemnq',
 934                 'ext': 'mp4',
 935                 'title': 'Violetta - Achter De Schermen - Ruggero',
 936                 'description': 'Achter de schermen met Ruggero',
 937                 'timestamp': 1435133761,
 938                 'upload_date': '20150624',
 939                 'uploader_id': 'echojecka',
 940             },
 941         },
 942         # Kaltura embed with single quotes
 943         {
 944             'url': 'http://fod.infobase.com/p_ViewPlaylist.aspx?AssignmentID=NUN8ZY',
 945             'info_dict': {
 946                 'id': '0_izeg5utt',
 947                 'ext': 'mp4',
 948                 'title': '35871',
 949                 'timestamp': 1355743100,
 950                 'upload_date': '20121217',
 951                 'uploader_id': 'batchUser',
 952             },
 953             'add_ie': ['Kaltura'],
 954         },
 955         {
 956             # Kaltura embedded via quoted entry_id
 957             'url': 'https://www.oreilly.com/ideas/my-cloud-makes-pretty-pictures',
 958             'info_dict': {
 959                 'id': '0_utuok90b',
 960                 'ext': 'mp4',
 961                 'title': '06_matthew_brender_raj_dutt',
 962                 'timestamp': 1466638791,
 963                 'upload_date': '20160622',
 964             },
 965             'add_ie': ['Kaltura'],
 966             'expected_warnings': [
 967                 'Could not send HEAD request'
 968             ],
 969             'params': {
 970                 'skip_download': True,
 971             }
 972         },
 973         # Eagle.Platform embed (generic URL)
 974         {
 975             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
 976             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
 977             'info_dict': {
 978                 'id': '227304',
 979                 'ext': 'mp4',
 980                 'title': 'Навальный вышел на свободу',
 981                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
 982                 'thumbnail': 're:^https?://.*\.jpg$',
 983                 'duration': 87,
 984                 'view_count': int,
 985                 'age_limit': 0,
 986             },
 987         },
 988         # ClipYou (Eagle.Platform) embed (custom URL)
 989         {
 990             'url': 'http://muz-tv.ru/play/7129/',
 991             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
 992             'info_dict': {
 993                 'id': '12820',
 994                 'ext': 'mp4',
 995                 'title': "'O Sole Mio",
 996                 'thumbnail': 're:^https?://.*\.jpg$',
 997                 'duration': 216,
 998                 'view_count': int,
 999             },
1000         },
1001         # Pladform embed
1002         {
1003             'url': 'http://muz-tv.ru/kinozal/view/7400/',
1004             'info_dict': {
1005                 'id': '100183293',
1006                 'ext': 'mp4',
1007                 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
1008                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
1009                 'thumbnail': 're:^https?://.*\.jpg$',
1010                 'duration': 694,
1011                 'age_limit': 0,
1012             },
1013         },
1014         # Playwire embed
1015         {
1016             'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
1017             'info_dict': {
1018                 'id': '3519514',
1019                 'ext': 'mp4',
1020                 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
1021                 'thumbnail': 're:^https?://.*\.png$',
1022                 'duration': 45.115,
1023             },
1024         },
1025         # 5min embed
1026         {
1027             'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
1028             'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
1029             'info_dict': {
1030                 'id': '518726732',
1031                 'ext': 'mp4',
1032                 'title': 'Facebook Creates "On This Day" | Crunch Report',
1033             },
1034         },
1035         # SVT embed
1036         {
1037             'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
1038             'info_dict': {
1039                 'id': '2900353',
1040                 'ext': 'flv',
1041                 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
1042                 'duration': 27,
1043                 'age_limit': 0,
1044             },
1045         },
1046         # Crooks and Liars embed
1047         {
1048             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
1049             'info_dict': {
1050                 'id': '8RUoRhRi',
1051                 'ext': 'mp4',
1052                 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
1053                 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
1054                 'timestamp': 1428207000,
1055                 'upload_date': '20150405',
1056                 'uploader': 'Heather',
1057             },
1058         },
1059         # Crooks and Liars external embed
1060         {
1061             'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
1062             'info_dict': {
1063                 'id': 'MTE3MjUtMzQ2MzA',
1064                 'ext': 'mp4',
1065                 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
1066                 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
1067                 'timestamp': 1265032391,
1068                 'upload_date': '20100201',
1069                 'uploader': 'Heather',
1070             },
1071         },
1072         # NBC Sports vplayer embed
1073         {
1074             'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
1075             'info_dict': {
1076                 'id': 'ln7x1qSThw4k',
1077                 'ext': 'flv',
1078                 'title': "PFT Live: New leader in the 'new-look' defense",
1079                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
1080                 'uploader': 'NBCU-SPORTS',
1081                 'upload_date': '20140107',
1082                 'timestamp': 1389118457,
1083             },
1084         },
1085         # NBC News embed
1086         {
1087             'url': 'http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html',
1088             'md5': '1aa589c675898ae6d37a17913cf68d66',
1089             'info_dict': {
1090                 'id': '701714499682',
1091                 'ext': 'mp4',
1092                 'title': 'PREVIEW: On Assignment: David Letterman',
1093                 'description': 'A preview of Tom Brokaw\'s interview with David Letterman as part of the On Assignment series powered by Dateline. Airs Sunday June 12 at 7/6c.',
1094             },
1095         },
1096         # UDN embed
1097         {
1098             'url': 'https://video.udn.com/news/300346',
1099             'md5': 'fd2060e988c326991037b9aff9df21a6',
1100             'info_dict': {
1101                 'id': '300346',
1102                 'ext': 'mp4',
1103                 'title': '中一中男師變性 全校師生力挺',
1104                 'thumbnail': 're:^https?://.*\.jpg$',
1105             },
1106             'params': {
1107                 # m3u8 download
1108                 'skip_download': True,
1109             },
1110         },
1111         # Ooyala embed
1112         {
1113             'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
1114             'info_dict': {
1115                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
1116                 'ext': 'mp4',
1117                 'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.',
1118                 'title': 'This is what separates the Excel masters from the wannabes',
1119                 'duration': 191.933,
1120             },
1121             'params': {
1122                 # m3u8 downloads
1123                 'skip_download': True,
1124             }
1125         },
1126         # Brightcove URL in single quotes
1127         {
1128             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
1129             'md5': '4ae374f1f8b91c889c4b9203c8c752af',
1130             'info_dict': {
1131                 'id': '4255764656001',
1132                 'ext': 'mp4',
1133                 'title': 'SN Presents: Russell Martin, World Citizen',
1134                 '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.',
1135                 'uploader': 'Rogers Sportsnet',
1136                 'uploader_id': '1704050871',
1137                 'upload_date': '20150525',
1138                 'timestamp': 1432570283,
1139             },
1140         },
1141         # Dailymotion Cloud video
1142         {
1143             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
1144             'md5': 'dcaf23ad0c67a256f4278bce6e0bae38',
1145             'info_dict': {
1146                 'id': 'x2uy8t3',
1147                 'ext': 'mp4',
1148                 'title': 'Sauvons les abeilles ! - Le débat',
1149                 'description': 'md5:d9082128b1c5277987825d684939ca26',
1150                 'thumbnail': 're:^https?://.*\.jpe?g$',
1151                 'timestamp': 1434970506,
1152                 'upload_date': '20150622',
1153                 'uploader': 'Public Sénat',
1154                 'uploader_id': 'xa9gza',
1155             }
1156         },
1157         # OnionStudios embed
1158         {
1159             'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
1160             'info_dict': {
1161                 'id': '2855',
1162                 'ext': 'mp4',
1163                 'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
1164                 'thumbnail': 're:^https?://.*\.jpe?g$',
1165                 'uploader': 'ClickHole',
1166                 'uploader_id': 'clickhole',
1167             }
1168         },
1169         # SnagFilms embed
1170         {
1171             'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
1172             'info_dict': {
1173                 'id': '74849a00-85a9-11e1-9660-123139220831',
1174                 'ext': 'mp4',
1175                 'title': '#whilewewatch',
1176             }
1177         },
1178         # AdobeTVVideo embed
1179         {
1180             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
1181             'md5': '43662b577c018ad707a63766462b1e87',
1182             'info_dict': {
1183                 'id': '2456',
1184                 'ext': 'mp4',
1185                 'title': 'New experience with Acrobat DC',
1186                 'description': 'New experience with Acrobat DC',
1187                 'duration': 248.667,
1188             },
1189         },
1190         # ScreenwaveMedia embed
1191         {
1192             'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1',
1193             'md5': '24ace5baba0d35d55c6810b51f34e9e0',
1194             'info_dict': {
1195                 'id': 'cinemasnob-55d26273809dd',
1196                 'ext': 'mp4',
1197                 'title': 'cinemasnob',
1198             },
1199         },
1200         # BrightcoveInPageEmbed embed
1201         {
1202             'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
1203             'info_dict': {
1204                 'id': '4238694884001',
1205                 'ext': 'flv',
1206                 'title': 'Tabletop: Dread, Last Thoughts',
1207                 'description': 'Tabletop: Dread, Last Thoughts',
1208                 'duration': 51690,
1209             },
1210         },
1211         # Brightcove embed, with no valid 'renditions' but valid 'IOSRenditions'
1212         # This video can't be played in browsers if Flash disabled and UA set to iPhone, which is actually a false alarm
1213         {
1214             'url': 'https://dl.dropboxusercontent.com/u/29092637/interview.html',
1215             'info_dict': {
1216                 'id': '4785848093001',
1217                 'ext': 'mp4',
1218                 'title': 'The Cardinal Pell Interview',
1219                 'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ',
1220                 'uploader': 'GlobeCast Australia - GlobeStream',
1221                 'uploader_id': '2733773828001',
1222                 'upload_date': '20160304',
1223                 'timestamp': 1457083087,
1224             },
1225             'params': {
1226                 # m3u8 downloads
1227                 'skip_download': True,
1228             },
1229         },
1230         # Another form of arte.tv embed
1231         {
1232             'url': 'http://www.tv-replay.fr/redirection/09-04-16/arte-reportage-arte-11508975.html',
1233             'md5': '850bfe45417ddf221288c88a0cffe2e2',
1234             'info_dict': {
1235                 'id': '030273-562_PLUS7-F',
1236                 'ext': 'mp4',
1237                 'title': 'ARTE Reportage - Nulle part, en France',
1238                 'description': 'md5:e3a0e8868ed7303ed509b9e3af2b870d',
1239                 'upload_date': '20160409',
1240             },
1241         },
1242         # LiveLeak embed
1243         {
1244             'url': 'http://www.wykop.pl/link/3088787/',
1245             'md5': 'ace83b9ed19b21f68e1b50e844fdf95d',
1246             'info_dict': {
1247                 'id': '874_1459135191',
1248                 'ext': 'mp4',
1249                 'title': 'Man shows poor quality of new apartment building',
1250                 'description': 'The wall is like a sand pile.',
1251                 'uploader': 'Lake8737',
1252             }
1253         },
1254         # Duplicated embedded video URLs
1255         {
1256             'url': 'http://www.hudl.com/athlete/2538180/highlights/149298443',
1257             'info_dict': {
1258                 'id': '149298443_480_16c25b74_2',
1259                 'ext': 'mp4',
1260                 'title': 'vs. Blue Orange Spring Game',
1261                 'uploader': 'www.hudl.com',
1262             },
1263         },
1264         # twitter:player:stream embed
1265         {
1266             'url': 'http://www.rtl.be/info/video/589263.aspx?CategoryID=288',
1267             'info_dict': {
1268                 'id': 'master',
1269                 'ext': 'mp4',
1270                 'title': 'Une nouvelle espèce de dinosaure découverte en Argentine',
1271                 'uploader': 'www.rtl.be',
1272             },
1273             'params': {
1274                 # m3u8 downloads
1275                 'skip_download': True,
1276             },
1277         },
1278         # twitter:player embed
1279         {
1280             'url': 'http://www.theatlantic.com/video/index/484130/what-do-black-holes-sound-like/',
1281             'md5': 'a3e0df96369831de324f0778e126653c',
1282             'info_dict': {
1283                 'id': '4909620399001',
1284                 'ext': 'mp4',
1285                 'title': 'What Do Black Holes Sound Like?',
1286                 'description': 'what do black holes sound like',
1287                 'upload_date': '20160524',
1288                 'uploader_id': '29913724001',
1289                 'timestamp': 1464107587,
1290                 'uploader': 'TheAtlantic',
1291             },
1292             'add_ie': ['BrightcoveLegacy'],
1293         },
1294         # Facebook <iframe> embed
1295         {
1296             'url': 'https://www.hostblogger.de/blog/archives/6181-Auto-jagt-Betonmischer.html',
1297             'md5': 'fbcde74f534176ecb015849146dd3aee',
1298             'info_dict': {
1299                 'id': '599637780109885',
1300                 'ext': 'mp4',
1301                 'title': 'Facebook video #599637780109885',
1302             },
1303         },
1304         # Facebook API embed
1305         {
1306             'url': 'http://www.lothype.com/blue-stars-2016-preview-standstill-full-show/',
1307             'md5': 'a47372ee61b39a7b90287094d447d94e',
1308             'info_dict': {
1309                 'id': '10153467542406923',
1310                 'ext': 'mp4',
1311                 'title': 'Facebook video #10153467542406923',
1312             },
1313         },
1314         # Wordpress "YouTube Video Importer" plugin
1315         {
1316             'url': 'http://www.lothype.com/blue-devils-drumline-stanford-lot-2016/',
1317             'md5': 'd16797741b560b485194eddda8121b48',
1318             'info_dict': {
1319                 'id': 'HNTXWDXV9Is',
1320                 'ext': 'mp4',
1321                 'title': 'Blue Devils Drumline Stanford lot 2016',
1322                 'upload_date': '20160627',
1323                 'uploader_id': 'GENOCIDE8GENERAL10',
1324                 'uploader': 'cylus cyrus',
1325             },
1326         },
1327         {
1328             # video stored on custom kaltura server
1329             'url': 'http://www.expansion.com/multimedia/videos.html?media=EQcM30NHIPv',
1330             'md5': '537617d06e64dfed891fa1593c4b30cc',
1331             'info_dict': {
1332                 'id': '0_1iotm5bh',
1333                 'ext': 'mp4',
1334                 'title': 'Elecciones británicas: 5 lecciones para Rajoy',
1335                 'description': 'md5:435a89d68b9760b92ce67ed227055f16',
1336                 'uploader_id': 'videos.expansion@el-mundo.net',
1337                 'upload_date': '20150429',
1338                 'timestamp': 1430303472,
1339             },
1340             'add_ie': ['Kaltura'],
1341         },
1342         {
1343             # Non-standard Vimeo embed
1344             'url': 'https://openclassrooms.com/courses/understanding-the-web',
1345             'md5': '64d86f1c7d369afd9a78b38cbb88d80a',
1346             'info_dict': {
1347                 'id': '148867247',
1348                 'ext': 'mp4',
1349                 'title': 'Understanding the web - Teaser',
1350                 'description': 'This is "Understanding the web - Teaser" by openclassrooms on Vimeo, the home for high quality videos and the people who love them.',
1351                 'upload_date': '20151214',
1352                 'uploader': 'OpenClassrooms',
1353                 'uploader_id': 'openclassrooms',
1354             },
1355             'add_ie': ['Vimeo'],
1356         },
1357         {
1358             # generic vimeo embed that requires original URL passed as Referer
1359             'url': 'http://racing4everyone.eu/2016/07/30/formula-1-2016-round12-germany/',
1360             'only_matching': True,
1361         },
1362         {
1363             'url': 'https://support.arkena.com/display/PLAY/Ways+to+embed+your+video',
1364             'md5': 'b96f2f71b359a8ecd05ce4e1daa72365',
1365             'info_dict': {
1366                 'id': 'b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe',
1367                 'ext': 'mp4',
1368                 'title': 'Big Buck Bunny',
1369                 'description': 'Royalty free test video',
1370                 'timestamp': 1432816365,
1371                 'upload_date': '20150528',
1372                 'is_live': False,
1373             },
1374             'params': {
1375                 'skip_download': True,
1376             },
1377             'add_ie': [ArkenaIE.ie_key()],
1378         },
1379         {
1380             'url': 'http://nova.bg/news/view/2016/08/16/156543/%D0%BD%D0%B0-%D0%BA%D0%BE%D1%81%D1%8A%D0%BC-%D0%BE%D1%82-%D0%B2%D0%B7%D1%80%D0%B8%D0%B2-%D0%BE%D1%82%D1%86%D0%B5%D0%BF%D0%B8%D1%85%D0%B0-%D1%86%D1%8F%D0%BB-%D0%BA%D0%B2%D0%B0%D1%80%D1%82%D0%B0%D0%BB-%D0%B7%D0%B0%D1%80%D0%B0%D0%B4%D0%B8-%D0%B8%D0%B7%D1%82%D0%B8%D1%87%D0%B0%D0%BD%D0%B5-%D0%BD%D0%B0-%D0%B3%D0%B0%D0%B7-%D0%B2-%D0%BF%D0%BB%D0%BE%D0%B2%D0%B4%D0%B8%D0%B2/',
1381             'info_dict': {
1382                 'id': '1c7141f46c',
1383                 'ext': 'mp4',
1384                 'title': 'НА КОСЪМ ОТ ВЗРИВ: Изтичане на газ на бензиностанция в Пловдив',
1385             },
1386             'params': {
1387                 'skip_download': True,
1388             },
1389             'add_ie': [Vbox7IE.ie_key()],
1390         },
1391         {
1392             # DBTV embeds
1393             'url': 'http://www.dagbladet.no/2016/02/23/nyheter/nordlys/ski/troms/ver/43254897/',
1394             'info_dict': {
1395                 'id': '43254897',
1396                 'title': 'Etter ett års planlegging, klaffet endelig alt: - Jeg måtte ta en liten dans',
1397             },
1398             'playlist_mincount': 3,
1399         },
1400         # {
1401         #     # TODO: find another test
1402         #     # http://schema.org/VideoObject
1403         #     'url': 'https://flipagram.com/f/nyvTSJMKId',
1404         #     'md5': '888dcf08b7ea671381f00fab74692755',
1405         #     'info_dict': {
1406         #         'id': 'nyvTSJMKId',
1407         #         'ext': 'mp4',
1408         #         'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction',
1409         #         'description': '#love for cats.',
1410         #         'timestamp': 1461244995,
1411         #         'upload_date': '20160421',
1412         #     },
1413         #     'params': {
1414         #         'force_generic_extractor': True,
1415         #     },
1416         # }
1417     ]
1418
1419     def report_following_redirect(self, new_url):
1420         """Report information extraction."""
1421         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
1422
1423     def _extract_rss(self, url, video_id, doc):
1424         playlist_title = doc.find('./channel/title').text
1425         playlist_desc_el = doc.find('./channel/description')
1426         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
1427
1428         entries = []
1429         for it in doc.findall('./channel/item'):
1430             next_url = xpath_text(it, 'link', fatal=False)
1431             if not next_url:
1432                 enclosure_nodes = it.findall('./enclosure')
1433                 for e in enclosure_nodes:
1434                     next_url = e.attrib.get('url')
1435                     if next_url:
1436                         break
1437
1438             if not next_url:
1439                 continue
1440
1441             entries.append({
1442                 '_type': 'url',
1443                 'url': next_url,
1444                 'title': it.find('title').text,
1445             })
1446
1447         return {
1448             '_type': 'playlist',
1449             'id': url,
1450             'title': playlist_title,
1451             'description': playlist_desc,
1452             'entries': entries,
1453         }
1454
1455     def _extract_camtasia(self, url, video_id, webpage):
1456         """ Returns None if no camtasia video can be found. """
1457
1458         camtasia_cfg = self._search_regex(
1459             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
1460             webpage, 'camtasia configuration file', default=None)
1461         if camtasia_cfg is None:
1462             return None
1463
1464         title = self._html_search_meta('DC.title', webpage, fatal=True)
1465
1466         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
1467         camtasia_cfg = self._download_xml(
1468             camtasia_url, video_id,
1469             note='Downloading camtasia configuration',
1470             errnote='Failed to download camtasia configuration')
1471         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
1472
1473         entries = []
1474         for n in fileset_node.getchildren():
1475             url_n = n.find('./uri')
1476             if url_n is None:
1477                 continue
1478
1479             entries.append({
1480                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
1481                 'title': '%s - %s' % (title, n.tag),
1482                 'url': compat_urlparse.urljoin(url, url_n.text),
1483                 'duration': float_or_none(n.find('./duration').text),
1484             })
1485
1486         return {
1487             '_type': 'playlist',
1488             'entries': entries,
1489             'title': title,
1490         }
1491
1492     def _real_extract(self, url):
1493         if url.startswith('//'):
1494             return {
1495                 '_type': 'url',
1496                 'url': self.http_scheme() + url,
1497             }
1498
1499         parsed_url = compat_urlparse.urlparse(url)
1500         if not parsed_url.scheme:
1501             default_search = self._downloader.params.get('default_search')
1502             if default_search is None:
1503                 default_search = 'fixup_error'
1504
1505             if default_search in ('auto', 'auto_warning', 'fixup_error'):
1506                 if '/' in url:
1507                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
1508                     return self.url_result('http://' + url)
1509                 elif default_search != 'fixup_error':
1510                     if default_search == 'auto_warning':
1511                         if re.match(r'^(?:url|URL)$', url):
1512                             raise ExtractorError(
1513                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
1514                                 expected=True)
1515                         else:
1516                             self._downloader.report_warning(
1517                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
1518                     return self.url_result('ytsearch:' + url)
1519
1520             if default_search in ('error', 'fixup_error'):
1521                 raise ExtractorError(
1522                     '%r is not a valid URL. '
1523                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
1524                     % (url, url), expected=True)
1525             else:
1526                 if ':' not in default_search:
1527                     default_search += ':'
1528                 return self.url_result(default_search + url)
1529
1530         url, smuggled_data = unsmuggle_url(url)
1531         force_videoid = None
1532         is_intentional = smuggled_data and smuggled_data.get('to_generic')
1533         if smuggled_data and 'force_videoid' in smuggled_data:
1534             force_videoid = smuggled_data['force_videoid']
1535             video_id = force_videoid
1536         else:
1537             video_id = self._generic_id(url)
1538
1539         self.to_screen('%s: Requesting header' % video_id)
1540
1541         head_req = HEADRequest(url)
1542         head_response = self._request_webpage(
1543             head_req, video_id,
1544             note=False, errnote='Could not send HEAD request to %s' % url,
1545             fatal=False)
1546
1547         if head_response is not False:
1548             # Check for redirect
1549             new_url = head_response.geturl()
1550             if url != new_url:
1551                 self.report_following_redirect(new_url)
1552                 if force_videoid:
1553                     new_url = smuggle_url(
1554                         new_url, {'force_videoid': force_videoid})
1555                 return self.url_result(new_url)
1556
1557         full_response = None
1558         if head_response is False:
1559             request = sanitized_Request(url)
1560             request.add_header('Accept-Encoding', '*')
1561             full_response = self._request_webpage(request, video_id)
1562             head_response = full_response
1563
1564         info_dict = {
1565             'id': video_id,
1566             'title': self._generic_title(url),
1567             'upload_date': unified_strdate(head_response.headers.get('Last-Modified'))
1568         }
1569
1570         # Check for direct link to a video
1571         content_type = head_response.headers.get('Content-Type', '').lower()
1572         m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>[^;\s]+)', content_type)
1573         if m:
1574             format_id = m.group('format_id')
1575             if format_id.endswith('mpegurl'):
1576                 formats = self._extract_m3u8_formats(url, video_id, 'mp4')
1577             elif format_id == 'f4m':
1578                 formats = self._extract_f4m_formats(url, video_id)
1579             else:
1580                 formats = [{
1581                     'format_id': m.group('format_id'),
1582                     'url': url,
1583                     'vcodec': 'none' if m.group('type') == 'audio' else None
1584                 }]
1585                 info_dict['direct'] = True
1586             self._sort_formats(formats)
1587             info_dict['formats'] = formats
1588             return info_dict
1589
1590         if not self._downloader.params.get('test', False) and not is_intentional:
1591             force = self._downloader.params.get('force_generic_extractor', False)
1592             self._downloader.report_warning(
1593                 '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
1594
1595         if not full_response:
1596             request = sanitized_Request(url)
1597             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
1598             # making it impossible to download only chunk of the file (yet we need only 512kB to
1599             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
1600             # that will always result in downloading the whole file that is not desirable.
1601             # Therefore for extraction pass we have to override Accept-Encoding to any in order
1602             # to accept raw bytes and being able to download only a chunk.
1603             # It may probably better to solve this by checking Content-Type for application/octet-stream
1604             # after HEAD request finishes, but not sure if we can rely on this.
1605             request.add_header('Accept-Encoding', '*')
1606             full_response = self._request_webpage(request, video_id)
1607
1608         first_bytes = full_response.read(512)
1609
1610         # Is it an M3U playlist?
1611         if first_bytes.startswith(b'#EXTM3U'):
1612             info_dict['formats'] = self._extract_m3u8_formats(url, video_id, 'mp4')
1613             self._sort_formats(info_dict['formats'])
1614             return info_dict
1615
1616         # Maybe it's a direct link to a video?
1617         # Be careful not to download the whole thing!
1618         if not is_html(first_bytes):
1619             self._downloader.report_warning(
1620                 'URL could be a direct video link, returning it as such.')
1621             info_dict.update({
1622                 'direct': True,
1623                 'url': url,
1624             })
1625             return info_dict
1626
1627         webpage = self._webpage_read_content(
1628             full_response, url, video_id, prefix=first_bytes)
1629
1630         self.report_extraction(video_id)
1631
1632         # Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest?
1633         try:
1634             doc = compat_etree_fromstring(webpage.encode('utf-8'))
1635             if doc.tag == 'rss':
1636                 return self._extract_rss(url, video_id, doc)
1637             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
1638                 smil = self._parse_smil(doc, url, video_id)
1639                 self._sort_formats(smil['formats'])
1640                 return smil
1641             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
1642                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
1643             elif re.match(r'(?i)^(?:{[^}]+})?MPD$', doc.tag):
1644                 info_dict['formats'] = self._parse_mpd_formats(
1645                     doc, video_id,
1646                     mpd_base_url=full_response.geturl().rpartition('/')[0],
1647                     mpd_url=url)
1648                 self._sort_formats(info_dict['formats'])
1649                 return info_dict
1650             elif re.match(r'^{http://ns\.adobe\.com/f4m/[12]\.0}manifest$', doc.tag):
1651                 info_dict['formats'] = self._parse_f4m_formats(doc, url, video_id)
1652                 self._sort_formats(info_dict['formats'])
1653                 return info_dict
1654         except compat_xml_parse_error:
1655             pass
1656
1657         # Is it a Camtasia project?
1658         camtasia_res = self._extract_camtasia(url, video_id, webpage)
1659         if camtasia_res is not None:
1660             return camtasia_res
1661
1662         # Sometimes embedded video player is hidden behind percent encoding
1663         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1664         # Unescaping the whole page allows to handle those cases in a generic way
1665         webpage = compat_urllib_parse_unquote(webpage)
1666
1667         # it's tempting to parse this further, but you would
1668         # have to take into account all the variations like
1669         #   Video Title - Site Name
1670         #   Site Name | Video Title
1671         #   Video Title - Tagline | Site Name
1672         # and so on and so forth; it's just not practical
1673         video_title = self._og_search_title(
1674             webpage, default=None) or self._html_search_regex(
1675             r'(?s)<title>(.*?)</title>', webpage, 'video title',
1676             default='video')
1677
1678         # Try to detect age limit automatically
1679         age_limit = self._rta_search(webpage)
1680         # And then there are the jokers who advertise that they use RTA,
1681         # but actually don't.
1682         AGE_LIMIT_MARKERS = [
1683             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1684         ]
1685         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1686             age_limit = 18
1687
1688         # video uploader is domain name
1689         video_uploader = self._search_regex(
1690             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1691
1692         video_description = self._og_search_description(webpage, default=None)
1693         video_thumbnail = self._og_search_thumbnail(webpage, default=None)
1694
1695         # Helper method
1696         def _playlist_from_matches(matches, getter=None, ie=None):
1697             urlrs = orderedSet(
1698                 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1699                 for m in matches)
1700             return self.playlist_result(
1701                 urlrs, playlist_id=video_id, playlist_title=video_title)
1702
1703         # Look for Brightcove Legacy Studio embeds
1704         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
1705         if bc_urls:
1706             self.to_screen('Brightcove video detected.')
1707             entries = [{
1708                 '_type': 'url',
1709                 'url': smuggle_url(bc_url, {'Referer': url}),
1710                 'ie_key': 'BrightcoveLegacy'
1711             } for bc_url in bc_urls]
1712
1713             return {
1714                 '_type': 'playlist',
1715                 'title': video_title,
1716                 'id': video_id,
1717                 'entries': entries,
1718             }
1719
1720         # Look for Brightcove New Studio embeds
1721         bc_urls = BrightcoveNewIE._extract_urls(webpage)
1722         if bc_urls:
1723             return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
1724
1725         # Look for ThePlatform embeds
1726         tp_urls = ThePlatformIE._extract_urls(webpage)
1727         if tp_urls:
1728             return _playlist_from_matches(tp_urls, ie='ThePlatform')
1729
1730         # Look for Vessel embeds
1731         vessel_urls = VesselIE._extract_urls(webpage)
1732         if vessel_urls:
1733             return _playlist_from_matches(vessel_urls, ie=VesselIE.ie_key())
1734
1735         # Look for embedded rtl.nl player
1736         matches = re.findall(
1737             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1738             webpage)
1739         if matches:
1740             return _playlist_from_matches(matches, ie='RtlNl')
1741
1742         vimeo_urls = VimeoIE._extract_urls(url, webpage)
1743         if vimeo_urls:
1744             return _playlist_from_matches(vimeo_urls, ie=VimeoIE.ie_key())
1745
1746         vid_me_embed_url = self._search_regex(
1747             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
1748             webpage, 'vid.me embed', default=None)
1749         if vid_me_embed_url is not None:
1750             return self.url_result(vid_me_embed_url, 'Vidme')
1751
1752         # Look for embedded YouTube player
1753         matches = re.findall(r'''(?x)
1754             (?:
1755                 <iframe[^>]+?src=|
1756                 data-video-url=|
1757                 <embed[^>]+?src=|
1758                 embedSWF\(?:\s*|
1759                 new\s+SWFObject\(
1760             )
1761             (["\'])
1762                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1763                 (?:embed|v|p)/.+?)
1764             \1''', webpage)
1765         if matches:
1766             return _playlist_from_matches(
1767                 matches, lambda m: unescapeHTML(m[1]))
1768
1769         # Look for lazyYT YouTube embed
1770         matches = re.findall(
1771             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1772         if matches:
1773             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1774
1775         # Look for Wordpress "YouTube Video Importer" plugin
1776         matches = re.findall(r'''(?x)<div[^>]+
1777             class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1778             data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1779         if matches:
1780             return _playlist_from_matches(matches, lambda m: m[-1])
1781
1782         matches = DailymotionIE._extract_urls(webpage)
1783         if matches:
1784             return _playlist_from_matches(matches)
1785
1786         # Look for embedded Dailymotion playlist player (#3822)
1787         m = re.search(
1788             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1789         if m:
1790             playlists = re.findall(
1791                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1792             if playlists:
1793                 return _playlist_from_matches(
1794                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1795
1796         # Look for embedded Wistia player
1797         match = re.search(
1798             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1799         if match:
1800             embed_url = self._proto_relative_url(
1801                 unescapeHTML(match.group('url')))
1802             return {
1803                 '_type': 'url_transparent',
1804                 'url': embed_url,
1805                 'ie_key': 'Wistia',
1806                 'uploader': video_uploader,
1807             }
1808
1809         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1810         if match:
1811             return {
1812                 '_type': 'url_transparent',
1813                 'url': 'wistia:%s' % match.group('id'),
1814                 'ie_key': 'Wistia',
1815                 'uploader': video_uploader,
1816             }
1817
1818         match = re.search(
1819             r'''(?sx)
1820                 <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*?
1821                 <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]+)\b.*?\2
1822             ''', webpage)
1823         if match:
1824             return self.url_result(self._proto_relative_url(
1825                 'wistia:%s' % match.group('id')), 'Wistia')
1826
1827         # Look for SVT player
1828         svt_url = SVTIE._extract_url(webpage)
1829         if svt_url:
1830             return self.url_result(svt_url, 'SVT')
1831
1832         # Look for embedded condenast player
1833         matches = re.findall(
1834             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1835             webpage)
1836         if matches:
1837             return {
1838                 '_type': 'playlist',
1839                 'entries': [{
1840                     '_type': 'url',
1841                     'ie_key': 'CondeNast',
1842                     'url': ma,
1843                 } for ma in matches],
1844                 'title': video_title,
1845                 'id': video_id,
1846             }
1847
1848         # Look for Bandcamp pages with custom domain
1849         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1850         if mobj is not None:
1851             burl = unescapeHTML(mobj.group(1))
1852             # Don't set the extractor because it can be a track url or an album
1853             return self.url_result(burl)
1854
1855         # Look for embedded Vevo player
1856         mobj = re.search(
1857             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1858         if mobj is not None:
1859             return self.url_result(mobj.group('url'))
1860
1861         # Look for embedded Viddler player
1862         mobj = re.search(
1863             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1864             webpage)
1865         if mobj is not None:
1866             return self.url_result(mobj.group('url'))
1867
1868         # Look for NYTimes player
1869         mobj = re.search(
1870             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1871             webpage)
1872         if mobj is not None:
1873             return self.url_result(mobj.group('url'))
1874
1875         # Look for Libsyn player
1876         mobj = re.search(
1877             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1878         if mobj is not None:
1879             return self.url_result(mobj.group('url'))
1880
1881         # Look for Ooyala videos
1882         mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1883                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1884                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1885                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1886         if mobj is not None:
1887             return OoyalaIE._build_url_result(smuggle_url(mobj.group('ec'), {'domain': url}))
1888
1889         # Look for multiple Ooyala embeds on SBN network websites
1890         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1891         if mobj is not None:
1892             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1893             if embeds:
1894                 return _playlist_from_matches(
1895                     embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
1896
1897         # Look for Aparat videos
1898         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1899         if mobj is not None:
1900             return self.url_result(mobj.group(1), 'Aparat')
1901
1902         # Look for MPORA videos
1903         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1904         if mobj is not None:
1905             return self.url_result(mobj.group(1), 'Mpora')
1906
1907         # Look for embedded NovaMov-based player
1908         mobj = re.search(
1909             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1910                     (?P<url>http://(?:(?:embed|www)\.)?
1911                         (?:novamov\.com|
1912                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
1913                            videoweed\.(?:es|com)|
1914                            movshare\.(?:net|sx|ag)|
1915                            divxstage\.(?:eu|net|ch|co|at|ag))
1916                         /embed\.php.+?)\1''', webpage)
1917         if mobj is not None:
1918             return self.url_result(mobj.group('url'))
1919
1920         # Look for embedded Facebook player
1921         facebook_url = FacebookIE._extract_url(webpage)
1922         if facebook_url is not None:
1923             return self.url_result(facebook_url, 'Facebook')
1924
1925         # Look for embedded VK player
1926         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1927         if mobj is not None:
1928             return self.url_result(mobj.group('url'), 'VK')
1929
1930         # Look for embedded Odnoklassniki player
1931         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1', webpage)
1932         if mobj is not None:
1933             return self.url_result(mobj.group('url'), 'Odnoklassniki')
1934
1935         # Look for embedded ivi player
1936         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1937         if mobj is not None:
1938             return self.url_result(mobj.group('url'), 'Ivi')
1939
1940         # Look for embedded Huffington Post player
1941         mobj = re.search(
1942             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1943         if mobj is not None:
1944             return self.url_result(mobj.group('url'), 'HuffPost')
1945
1946         # Look for embed.ly
1947         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1948         if mobj is not None:
1949             return self.url_result(mobj.group('url'))
1950         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1951         if mobj is not None:
1952             return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
1953
1954         # Look for funnyordie embed
1955         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1956         if matches:
1957             return _playlist_from_matches(
1958                 matches, getter=unescapeHTML, ie='FunnyOrDie')
1959
1960         # Look for BBC iPlayer embed
1961         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1962         if matches:
1963             return _playlist_from_matches(matches, ie='BBCCoUk')
1964
1965         # Look for embedded RUTV player
1966         rutv_url = RUTVIE._extract_url(webpage)
1967         if rutv_url:
1968             return self.url_result(rutv_url, 'RUTV')
1969
1970         # Look for embedded TVC player
1971         tvc_url = TVCIE._extract_url(webpage)
1972         if tvc_url:
1973             return self.url_result(tvc_url, 'TVC')
1974
1975         # Look for embedded SportBox player
1976         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1977         if sportbox_urls:
1978             return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1979
1980         # Look for embedded PornHub player
1981         pornhub_url = PornHubIE._extract_url(webpage)
1982         if pornhub_url:
1983             return self.url_result(pornhub_url, 'PornHub')
1984
1985         # Look for embedded XHamster player
1986         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
1987         if xhamster_urls:
1988             return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
1989
1990         # Look for embedded TNAFlixNetwork player
1991         tnaflix_urls = TNAFlixNetworkEmbedIE._extract_urls(webpage)
1992         if tnaflix_urls:
1993             return _playlist_from_matches(tnaflix_urls, ie=TNAFlixNetworkEmbedIE.ie_key())
1994
1995         # Look for embedded Tvigle player
1996         mobj = re.search(
1997             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
1998         if mobj is not None:
1999             return self.url_result(mobj.group('url'), 'Tvigle')
2000
2001         # Look for embedded TED player
2002         mobj = re.search(
2003             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
2004         if mobj is not None:
2005             return self.url_result(mobj.group('url'), 'TED')
2006
2007         # Look for embedded Ustream videos
2008         mobj = re.search(
2009             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
2010         if mobj is not None:
2011             return self.url_result(mobj.group('url'), 'Ustream')
2012
2013         # Look for embedded arte.tv player
2014         mobj = re.search(
2015             r'<(?:script|iframe) [^>]*?src="(?P<url>http://www\.arte\.tv/(?:playerv2/embed|arte_vp/index)[^"]+)"',
2016             webpage)
2017         if mobj is not None:
2018             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
2019
2020         # Look for embedded francetv player
2021         mobj = re.search(
2022             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
2023             webpage)
2024         if mobj is not None:
2025             return self.url_result(mobj.group('url'))
2026
2027         # Look for embedded smotri.com player
2028         smotri_url = SmotriIE._extract_url(webpage)
2029         if smotri_url:
2030             return self.url_result(smotri_url, 'Smotri')
2031
2032         # Look for embedded Myvi.ru player
2033         myvi_url = MyviIE._extract_url(webpage)
2034         if myvi_url:
2035             return self.url_result(myvi_url)
2036
2037         # Look for embedded soundcloud player
2038         soundcloud_urls = SoundcloudIE._extract_urls(webpage)
2039         if soundcloud_urls:
2040             return _playlist_from_matches(soundcloud_urls, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
2041
2042         # Look for embedded mtvservices player
2043         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
2044         if mtvservices_url:
2045             return self.url_result(mtvservices_url, ie='MTVServicesEmbedded')
2046
2047         # Look for embedded yahoo player
2048         mobj = re.search(
2049             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
2050             webpage)
2051         if mobj is not None:
2052             return self.url_result(mobj.group('url'), 'Yahoo')
2053
2054         # Look for embedded sbs.com.au player
2055         mobj = re.search(
2056             r'''(?x)
2057             (?:
2058                 <meta\s+property="og:video"\s+content=|
2059                 <iframe[^>]+?src=
2060             )
2061             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
2062             webpage)
2063         if mobj is not None:
2064             return self.url_result(mobj.group('url'), 'SBS')
2065
2066         # Look for embedded Cinchcast player
2067         mobj = re.search(
2068             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
2069             webpage)
2070         if mobj is not None:
2071             return self.url_result(mobj.group('url'), 'Cinchcast')
2072
2073         mobj = re.search(
2074             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
2075             webpage)
2076         if not mobj:
2077             mobj = re.search(
2078                 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
2079                 webpage)
2080         if mobj is not None:
2081             return self.url_result(mobj.group('url'), 'MLB')
2082
2083         mobj = re.search(
2084             r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
2085             webpage)
2086         if mobj is not None:
2087             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
2088
2089         mobj = re.search(
2090             r'<iframe[^>]+src="(?P<url>https?://(?:new\.)?livestream\.com/[^"]+/player[^"]+)"',
2091             webpage)
2092         if mobj is not None:
2093             return self.url_result(mobj.group('url'), 'Livestream')
2094
2095         # Look for Zapiks embed
2096         mobj = re.search(
2097             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
2098         if mobj is not None:
2099             return self.url_result(mobj.group('url'), 'Zapiks')
2100
2101         # Look for Kaltura embeds
2102         kaltura_url = KalturaIE._extract_url(webpage)
2103         if kaltura_url:
2104             return self.url_result(smuggle_url(kaltura_url, {'source_url': url}), KalturaIE.ie_key())
2105
2106         # Look for Eagle.Platform embeds
2107         eagleplatform_url = EaglePlatformIE._extract_url(webpage)
2108         if eagleplatform_url:
2109             return self.url_result(eagleplatform_url, EaglePlatformIE.ie_key())
2110
2111         # Look for ClipYou (uses Eagle.Platform) embeds
2112         mobj = re.search(
2113             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
2114         if mobj is not None:
2115             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
2116
2117         # Look for Pladform embeds
2118         pladform_url = PladformIE._extract_url(webpage)
2119         if pladform_url:
2120             return self.url_result(pladform_url)
2121
2122         # Look for Videomore embeds
2123         videomore_url = VideomoreIE._extract_url(webpage)
2124         if videomore_url:
2125             return self.url_result(videomore_url)
2126
2127         # Look for Playwire embeds
2128         mobj = re.search(
2129             r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
2130         if mobj is not None:
2131             return self.url_result(mobj.group('url'))
2132
2133         # Look for 5min embeds
2134         mobj = re.search(
2135             r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
2136         if mobj is not None:
2137             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
2138
2139         # Look for Crooks and Liars embeds
2140         mobj = re.search(
2141             r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
2142         if mobj is not None:
2143             return self.url_result(mobj.group('url'))
2144
2145         # Look for NBC Sports VPlayer embeds
2146         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
2147         if nbc_sports_url:
2148             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
2149
2150         # Look for NBC News embeds
2151         nbc_news_embed_url = re.search(
2152             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//www\.nbcnews\.com/widget/video-embed/[^"\']+)\1', webpage)
2153         if nbc_news_embed_url:
2154             return self.url_result(nbc_news_embed_url.group('url'), 'NBCNews')
2155
2156         # Look for Google Drive embeds
2157         google_drive_url = GoogleDriveIE._extract_url(webpage)
2158         if google_drive_url:
2159             return self.url_result(google_drive_url, 'GoogleDrive')
2160
2161         # Look for UDN embeds
2162         mobj = re.search(
2163             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage)
2164         if mobj is not None:
2165             return self.url_result(
2166                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
2167
2168         # Look for Senate ISVP iframe
2169         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
2170         if senate_isvp_url:
2171             return self.url_result(senate_isvp_url, 'SenateISVP')
2172
2173         # Look for Dailymotion Cloud videos
2174         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
2175         if dmcloud_url:
2176             return self.url_result(dmcloud_url, 'DailymotionCloud')
2177
2178         # Look for OnionStudios embeds
2179         onionstudios_url = OnionStudiosIE._extract_url(webpage)
2180         if onionstudios_url:
2181             return self.url_result(onionstudios_url)
2182
2183         # Look for ViewLift embeds
2184         viewlift_url = ViewLiftEmbedIE._extract_url(webpage)
2185         if viewlift_url:
2186             return self.url_result(viewlift_url)
2187
2188         # Look for JWPlatform embeds
2189         jwplatform_url = JWPlatformIE._extract_url(webpage)
2190         if jwplatform_url:
2191             return self.url_result(jwplatform_url, 'JWPlatform')
2192
2193         # Look for ScreenwaveMedia embeds
2194         mobj = re.search(ScreenwaveMediaIE.EMBED_PATTERN, webpage)
2195         if mobj is not None:
2196             return self.url_result(unescapeHTML(mobj.group('url')), 'ScreenwaveMedia')
2197
2198         # Look for Digiteka embeds
2199         digiteka_url = DigitekaIE._extract_url(webpage)
2200         if digiteka_url:
2201             return self.url_result(self._proto_relative_url(digiteka_url), DigitekaIE.ie_key())
2202
2203         # Look for Arkena embeds
2204         arkena_url = ArkenaIE._extract_url(webpage)
2205         if arkena_url:
2206             return self.url_result(arkena_url, ArkenaIE.ie_key())
2207
2208         # Look for Limelight embeds
2209         mobj = re.search(r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})', webpage)
2210         if mobj:
2211             lm = {
2212                 'Media': 'media',
2213                 'Channel': 'channel',
2214                 'ChannelList': 'channel_list',
2215             }
2216             return self.url_result('limelight:%s:%s' % (
2217                 lm[mobj.group(1)], mobj.group(2)), 'Limelight%s' % mobj.group(1), mobj.group(2))
2218
2219         # Look for AdobeTVVideo embeds
2220         mobj = re.search(
2221             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
2222             webpage)
2223         if mobj is not None:
2224             return self.url_result(
2225                 self._proto_relative_url(unescapeHTML(mobj.group(1))),
2226                 'AdobeTVVideo')
2227
2228         # Look for Vine embeds
2229         mobj = re.search(
2230             r'<iframe[^>]+src=[\'"]((?:https?:)?//(?:www\.)?vine\.co/v/[^/]+/embed/(?:simple|postcard))',
2231             webpage)
2232         if mobj is not None:
2233             return self.url_result(
2234                 self._proto_relative_url(unescapeHTML(mobj.group(1))), 'Vine')
2235
2236         # Look for VODPlatform embeds
2237         mobj = re.search(
2238             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vod-platform\.net/[eE]mbed/.+?)\1',
2239             webpage)
2240         if mobj is not None:
2241             return self.url_result(
2242                 self._proto_relative_url(unescapeHTML(mobj.group('url'))), 'VODPlatform')
2243
2244         # Look for Mangomolo embeds
2245         mobj = re.search(
2246             r'''(?x)<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?admin\.mangomolo\.com/analytics/index\.php/customers/embed/
2247                 (?:
2248                     video\?.*?\bid=(?P<video_id>\d+)|
2249                     index\?.*?\bchannelid=(?P<channel_id>(?:[A-Za-z0-9+/=]|%2B|%2F|%3D)+)
2250                 ).+?)\1''', webpage)
2251         if mobj is not None:
2252             info = {
2253                 '_type': 'url_transparent',
2254                 'url': self._proto_relative_url(unescapeHTML(mobj.group('url'))),
2255                 'title': video_title,
2256                 'description': video_description,
2257                 'thumbnail': video_thumbnail,
2258                 'uploader': video_uploader,
2259             }
2260             video_id = mobj.group('video_id')
2261             if video_id:
2262                 info.update({
2263                     'ie_key': 'MangomoloVideo',
2264                     'id': video_id,
2265                 })
2266             else:
2267                 info.update({
2268                     'ie_key': 'MangomoloLive',
2269                     'id': mobj.group('channel_id'),
2270                 })
2271             return info
2272
2273         # Look for Instagram embeds
2274         instagram_embed_url = InstagramIE._extract_embed_url(webpage)
2275         if instagram_embed_url is not None:
2276             return self.url_result(
2277                 self._proto_relative_url(instagram_embed_url), InstagramIE.ie_key())
2278
2279         # Look for LiveLeak embeds
2280         liveleak_url = LiveLeakIE._extract_url(webpage)
2281         if liveleak_url:
2282             return self.url_result(liveleak_url, 'LiveLeak')
2283
2284         # Look for 3Q SDN embeds
2285         threeqsdn_url = ThreeQSDNIE._extract_url(webpage)
2286         if threeqsdn_url:
2287             return {
2288                 '_type': 'url_transparent',
2289                 'ie_key': ThreeQSDNIE.ie_key(),
2290                 'url': self._proto_relative_url(threeqsdn_url),
2291                 'title': video_title,
2292                 'description': video_description,
2293                 'thumbnail': video_thumbnail,
2294                 'uploader': video_uploader,
2295             }
2296
2297         # Look for VBOX7 embeds
2298         vbox7_url = Vbox7IE._extract_url(webpage)
2299         if vbox7_url:
2300             return self.url_result(vbox7_url, Vbox7IE.ie_key())
2301
2302         # Look for DBTV embeds
2303         dbtv_urls = DBTVIE._extract_urls(webpage)
2304         if dbtv_urls:
2305             return _playlist_from_matches(dbtv_urls, ie=DBTVIE.ie_key())
2306
2307         # Looking for http://schema.org/VideoObject
2308         json_ld = self._search_json_ld(
2309             webpage, video_id, default={}, expected_type='VideoObject')
2310         if json_ld.get('url'):
2311             info_dict.update({
2312                 'title': video_title or info_dict['title'],
2313                 'description': video_description,
2314                 'thumbnail': video_thumbnail,
2315                 'age_limit': age_limit
2316             })
2317             info_dict.update(json_ld)
2318             return info_dict
2319
2320         # Look for HTML5 media
2321         entries = self._parse_html5_media_entries(url, webpage, video_id, m3u8_id='hls')
2322         if entries:
2323             for entry in entries:
2324                 entry.update({
2325                     'id': video_id,
2326                     'title': video_title,
2327                 })
2328                 self._sort_formats(entry['formats'])
2329             return self.playlist_result(entries)
2330
2331         def check_video(vurl):
2332             if YoutubeIE.suitable(vurl):
2333                 return True
2334             vpath = compat_urlparse.urlparse(vurl).path
2335             vext = determine_ext(vpath)
2336             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml', 'js')
2337
2338         def filter_video(urls):
2339             return list(filter(check_video, urls))
2340
2341         # Start with something easy: JW Player in SWFObject
2342         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
2343         if not found:
2344             # Look for gorilla-vid style embedding
2345             found = filter_video(re.findall(r'''(?sx)
2346                 (?:
2347                     jw_plugins|
2348                     JWPlayerOptions|
2349                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
2350                 )
2351                 .*?
2352                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
2353         if not found:
2354             # Broaden the search a little bit
2355             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
2356         if not found:
2357             # Broaden the findall a little bit: JWPlayer JS loader
2358             found = filter_video(re.findall(
2359                 r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
2360         if not found:
2361             # Flow player
2362             found = filter_video(re.findall(r'''(?xs)
2363                 flowplayer\("[^"]+",\s*
2364                     \{[^}]+?\}\s*,
2365                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
2366                         ["']?url["']?\s*:\s*["']([^"']+)["']
2367             ''', webpage))
2368         if not found:
2369             # Cinerama player
2370             found = re.findall(
2371                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
2372         if not found:
2373             # Try to find twitter cards info
2374             # twitter:player:stream should be checked before twitter:player since
2375             # it is expected to contain a raw stream (see
2376             # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser)
2377             found = filter_video(re.findall(
2378                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
2379         if not found:
2380             # We look for Open Graph info:
2381             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
2382             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
2383             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
2384             if m_video_type is not None:
2385                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
2386         if not found:
2387             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
2388             found = re.search(
2389                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
2390                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
2391                 webpage)
2392             if not found:
2393                 # Look also in Refresh HTTP header
2394                 refresh_header = head_response.headers.get('Refresh')
2395                 if refresh_header:
2396                     # In python 2 response HTTP headers are bytestrings
2397                     if sys.version_info < (3, 0) and isinstance(refresh_header, str):
2398                         refresh_header = refresh_header.decode('iso-8859-1')
2399                     found = re.search(REDIRECT_REGEX, refresh_header)
2400             if found:
2401                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
2402                 self.report_following_redirect(new_url)
2403                 return {
2404                     '_type': 'url',
2405                     'url': new_url,
2406                 }
2407
2408         if not found:
2409             # twitter:player is a https URL to iframe player that may or may not
2410             # be supported by youtube-dl thus this is checked the very last (see
2411             # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser)
2412             embed_url = self._html_search_meta('twitter:player', webpage, default=None)
2413             if embed_url:
2414                 return self.url_result(embed_url)
2415
2416         if not found:
2417             raise UnsupportedError(url)
2418
2419         entries = []
2420         for video_url in orderedSet(found):
2421             video_url = unescapeHTML(video_url)
2422             video_url = video_url.replace('\\/', '/')
2423             video_url = compat_urlparse.urljoin(url, video_url)
2424             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
2425
2426             # Sometimes, jwplayer extraction will result in a YouTube URL
2427             if YoutubeIE.suitable(video_url):
2428                 entries.append(self.url_result(video_url, 'Youtube'))
2429                 continue
2430
2431             # here's a fun little line of code for you:
2432             video_id = os.path.splitext(video_id)[0]
2433
2434             entry_info_dict = {
2435                 'id': video_id,
2436                 'uploader': video_uploader,
2437                 'title': video_title,
2438                 'age_limit': age_limit,
2439             }
2440
2441             ext = determine_ext(video_url)
2442             if ext == 'smil':
2443                 entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
2444             elif ext == 'xspf':
2445                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
2446             elif ext == 'm3u8':
2447                 entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
2448             elif ext == 'mpd':
2449                 entry_info_dict['formats'] = self._extract_mpd_formats(video_url, video_id)
2450             elif ext == 'f4m':
2451                 entry_info_dict['formats'] = self._extract_f4m_formats(video_url, video_id)
2452             else:
2453                 entry_info_dict['url'] = video_url
2454
2455             if entry_info_dict.get('formats'):
2456                 self._sort_formats(entry_info_dict['formats'])
2457
2458             entries.append(entry_info_dict)
2459
2460         if len(entries) == 1:
2461             return entries[0]
2462         else:
2463             for num, e in enumerate(entries, start=1):
2464                 # 'url' results don't have a title
2465                 if e.get('title') is not None:
2466                     e['title'] = '%s (%d)' % (e['title'], num)
2467             return {
2468                 '_type': 'playlist',
2469                 'entries': entries,
2470             }