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

   1 # encoding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 import os
   6 import re
   7 import sys
   8
   9 from .common import InfoExtractor
  10 from .youtube import YoutubeIE
  11 from ..compat import (
  12     compat_etree_fromstring,
  13     compat_urllib_parse_unquote,
  14     compat_urlparse,
  15     compat_xml_parse_error,
  16 )
  17 from ..utils import (
  18     determine_ext,
  19     ExtractorError,
  20     float_or_none,
  21     HEADRequest,
  22     is_html,
  23     orderedSet,
  24     sanitized_Request,
  25     smuggle_url,
  26     unescapeHTML,
  27     unified_strdate,
  28     unsmuggle_url,
  29     UnsupportedError,
  30     url_basename,
  31     xpath_text,
  32 )
  33 from .brightcove import (
  34     BrightcoveLegacyIE,
  35     BrightcoveNewIE,
  36 )
  37 from .nbc import NBCSportsVPlayerIE
  38 from .ooyala import OoyalaIE
  39 from .rutv import RUTVIE
  40 from .tvc import TVCIE
  41 from .sportbox import SportBoxEmbedIE
  42 from .smotri import SmotriIE
  43 from .myvi import MyviIE
  44 from .condenast import CondeNastIE
  45 from .udn import UDNEmbedIE
  46 from .senateisvp import SenateISVPIE
  47 from .svt import SVTIE
  48 from .pornhub import PornHubIE
  49 from .xhamster import XHamsterEmbedIE
  50 from .tnaflix import TNAFlixNetworkEmbedIE
  51 from .vimeo import VimeoIE
  52 from .dailymotion import (
  53     DailymotionIE,
  54     DailymotionCloudIE,
  55 )
  56 from .onionstudios import OnionStudiosIE
  57 from .viewlift import ViewLiftEmbedIE
  58 from .screenwavemedia import ScreenwaveMediaIE
  59 from .mtv import MTVServicesEmbeddedIE
  60 from .pladform import PladformIE
  61 from .videomore import VideomoreIE
  62 from .googledrive import GoogleDriveIE
  63 from .jwplatform import JWPlatformIE
  64 from .digiteka import DigitekaIE
  65 from .arkena import ArkenaIE
  66 from .instagram import InstagramIE
  67 from .liveleak import LiveLeakIE
  68 from .threeqsdn import ThreeQSDNIE
  69 from .theplatform import ThePlatformIE
  70 from .vessel import VesselIE
  71 from .kaltura import KalturaIE
  72 from .eagleplatform import EaglePlatformIE
  73 from .facebook import FacebookIE
  74 from .soundcloud import SoundcloudIE
  75 from .vbox7 import Vbox7IE
  76 from .dbtv import DBTVIE
  77
  78
  79 class GenericIE(InfoExtractor):
  80     IE_DESC = 'Generic downloader that works on some sites'
  81     _VALID_URL = r'.*'
  82     IE_NAME = 'generic'
  83     _TESTS = [
  84         # Direct link to a video
  85         {
  86             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  87             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  88             'info_dict': {
  89                 'id': 'trailer',
  90                 'ext': 'mp4',
  91                 'title': 'trailer',
  92                 'upload_date': '20100513',
  93             }
  94         },
  95         # Direct link to media delivered compressed (until Accept-Encoding is *)
  96         {
  97             'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
  98             'md5': '128c42e68b13950268b648275386fc74',
  99             'info_dict': {
 100                 'id': 'FictionJunction-Parallel_Hearts',
 101                 'ext': 'flac',
 102                 'title': 'FictionJunction-Parallel_Hearts',
 103                 'upload_date': '20140522',
 104             },
 105             'expected_warnings': [
 106                 'URL could be a direct video link, returning it as such.'
 107             ],
 108             'skip': 'URL invalid',
 109         },
 110         # Direct download with broken HEAD
 111         {
 112             'url': 'http://ai-radio.org:8000/radio.opus',
 113             'info_dict': {
 114                 'id': 'radio',
 115                 'ext': 'opus',
 116                 'title': 'radio',
 117             },
 118             'params': {
 119                 'skip_download': True,  # infinite live stream
 120             },
 121             'expected_warnings': [
 122                 r'501.*Not Implemented',
 123                 r'400.*Bad Request',
 124             ],
 125         },
 126         # Direct link with incorrect MIME type
 127         {
 128             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 129             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 130             'info_dict': {
 131                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 132                 'id': '5_Lennart_Poettering_-_Systemd',
 133                 'ext': 'webm',
 134                 'title': '5_Lennart_Poettering_-_Systemd',
 135                 'upload_date': '20141120',
 136             },
 137             'expected_warnings': [
 138                 'URL could be a direct video link, returning it as such.'
 139             ]
 140         },
 141         # RSS feed
 142         {
 143             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 144             'info_dict': {
 145                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 146                 'title': 'Zero Punctuation',
 147                 'description': 're:.*groundbreaking video review series.*'
 148             },
 149             'playlist_mincount': 11,
 150         },
 151         # RSS feed with enclosure
 152         {
 153             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 154             'info_dict': {
 155                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 156                 'ext': 'm4v',
 157                 'upload_date': '20150228',
 158                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 159             }
 160         },
 161         # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
 162         {
 163             'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
 164             'info_dict': {
 165                 'id': 'smil',
 166                 'ext': 'mp4',
 167                 'title': 'Automatics, robotics and biocybernetics',
 168                 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
 169                 'upload_date': '20130627',
 170                 'formats': 'mincount:16',
 171                 'subtitles': 'mincount:1',
 172             },
 173             'params': {
 174                 'force_generic_extractor': True,
 175                 'skip_download': True,
 176             },
 177         },
 178         # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html
 179         {
 180             'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil',
 181             'info_dict': {
 182                 'id': 'hds',
 183                 'ext': 'flv',
 184                 'title': 'hds',
 185                 'formats': 'mincount:1',
 186             },
 187             'params': {
 188                 'skip_download': True,
 189             },
 190         },
 191         # SMIL from https://www.restudy.dk/video/play/id/1637
 192         {
 193             'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml',
 194             'info_dict': {
 195                 'id': 'video_1637',
 196                 'ext': 'flv',
 197                 'title': 'video_1637',
 198                 'formats': 'mincount:3',
 199             },
 200             'params': {
 201                 'skip_download': True,
 202             },
 203         },
 204         # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm
 205         {
 206             'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil',
 207             'info_dict': {
 208                 'id': 'smil-service',
 209                 'ext': 'flv',
 210                 'title': 'smil-service',
 211                 'formats': 'mincount:1',
 212             },
 213             'params': {
 214                 'skip_download': True,
 215             },
 216         },
 217         # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370
 218         {
 219             'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil',
 220             'info_dict': {
 221                 'id': '4719370',
 222                 'ext': 'mp4',
 223                 'title': '571de1fd-47bc-48db-abf9-238872a58d1f',
 224                 'formats': 'mincount:3',
 225             },
 226             'params': {
 227                 'skip_download': True,
 228             },
 229         },
 230         # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html
 231         {
 232             'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf',
 233             'info_dict': {
 234                 'id': 'mZlp2ctYIUEB',
 235                 'ext': 'mp4',
 236                 'title': 'Tikibad ontruimd wegens brand',
 237                 'description': 'md5:05ca046ff47b931f9b04855015e163a4',
 238                 'thumbnail': 're:^https?://.*\.jpg$',
 239                 'duration': 33,
 240             },
 241             'params': {
 242                 'skip_download': True,
 243             },
 244         },
 245         # MPD from http://dash-mse-test.appspot.com/media.html
 246         {
 247             'url': 'http://yt-dash-mse-test.commondatastorage.googleapis.com/media/car-20120827-manifest.mpd',
 248             'md5': '4b57baab2e30d6eb3a6a09f0ba57ef53',
 249             'info_dict': {
 250                 'id': 'car-20120827-manifest',
 251                 'ext': 'mp4',
 252                 'title': 'car-20120827-manifest',
 253                 'formats': 'mincount:9',
 254                 'upload_date': '20130904',
 255             },
 256             'params': {
 257                 'format': 'bestvideo',
 258             },
 259         },
 260         # m3u8 served with Content-Type: audio/x-mpegURL; charset=utf-8
 261         {
 262             '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',
 263             'info_dict': {
 264                 'id': 'content',
 265                 'ext': 'mp4',
 266                 'title': 'content',
 267                 'formats': 'mincount:8',
 268             },
 269             'params': {
 270                 # m3u8 downloads
 271                 'skip_download': True,
 272             },
 273             'skip': 'video gone',
 274         },
 275         # m3u8 served with Content-Type: text/plain
 276         {
 277             'url': 'http://www.nacentapps.com/m3u8/index.m3u8',
 278             'info_dict': {
 279                 'id': 'index',
 280                 'ext': 'mp4',
 281                 'title': 'index',
 282                 'upload_date': '20140720',
 283                 'formats': 'mincount:11',
 284             },
 285             'params': {
 286                 # m3u8 downloads
 287                 'skip_download': True,
 288             },
 289             'skip': 'video gone',
 290         },
 291         # google redirect
 292         {
 293             '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',
 294             'info_dict': {
 295                 'id': 'cmQHVoWB5FY',
 296                 'ext': 'mp4',
 297                 'upload_date': '20130224',
 298                 'uploader_id': 'TheVerge',
 299                 'description': 're:^Chris Ziegler takes a look at the\.*',
 300                 'uploader': 'The Verge',
 301                 'title': 'First Firefox OS phones side-by-side',
 302             },
 303             'params': {
 304                 'skip_download': False,
 305             }
 306         },
 307         {
 308             # redirect in Refresh HTTP header
 309             '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',
 310             'info_dict': {
 311                 'id': 'pO8h3EaFRdo',
 312                 'ext': 'mp4',
 313                 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
 314                 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5',
 315                 'upload_date': '20150917',
 316                 'uploader_id': 'brtvofficial',
 317                 'uploader': 'Boiler Room',
 318             },
 319             'params': {
 320                 'skip_download': False,
 321             },
 322         },
 323         {
 324             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
 325             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
 326             'info_dict': {
 327                 'id': '13601338388002',
 328                 'ext': 'mp4',
 329                 'uploader': 'www.hodiho.fr',
 330                 'title': 'R\u00e9gis plante sa Jeep',
 331             }
 332         },
 333         # bandcamp page with custom domain
 334         {
 335             'add_ie': ['Bandcamp'],
 336             'url': 'http://bronyrock.com/track/the-pony-mash',
 337             'info_dict': {
 338                 'id': '3235767654',
 339                 'ext': 'mp3',
 340                 'title': 'The Pony Mash',
 341                 'uploader': 'M_Pallante',
 342             },
 343             'skip': 'There is a limit of 200 free downloads / month for the test song',
 344         },
 345         # embedded brightcove video
 346         # it also tests brightcove videos that need to set the 'Referer' in the
 347         # http requests
 348         {
 349             'add_ie': ['BrightcoveLegacy'],
 350             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
 351             'info_dict': {
 352                 'id': '2765128793001',
 353                 'ext': 'mp4',
 354                 'title': 'Le cours de bourse : l’analyse technique',
 355                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
 356                 'uploader': 'BFM BUSINESS',
 357             },
 358             'params': {
 359                 'skip_download': True,
 360             },
 361         },
 362         {
 363             # https://github.com/rg3/youtube-dl/issues/2253
 364             'url': 'http://bcove.me/i6nfkrc3',
 365             'md5': '0ba9446db037002366bab3b3eb30c88c',
 366             'info_dict': {
 367                 'id': '3101154703001',
 368                 'ext': 'mp4',
 369                 'title': 'Still no power',
 370                 'uploader': 'thestar.com',
 371                 '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.',
 372             },
 373             'add_ie': ['BrightcoveLegacy'],
 374             'skip': 'video gone',
 375         },
 376         {
 377             'url': 'http://www.championat.com/video/football/v/87/87499.html',
 378             'md5': 'fb973ecf6e4a78a67453647444222983',
 379             'info_dict': {
 380                 'id': '3414141473001',
 381                 'ext': 'mp4',
 382                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 383                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 384                 'uploader': 'Championat',
 385             },
 386         },
 387         {
 388             # https://github.com/rg3/youtube-dl/issues/3541
 389             'add_ie': ['BrightcoveLegacy'],
 390             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 391             'info_dict': {
 392                 'id': '3866516442001',
 393                 'ext': 'mp4',
 394                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 395                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 396                 'uploader': 'SBS Broadcasting',
 397             },
 398             'skip': 'Restricted to Netherlands',
 399             'params': {
 400                 'skip_download': True,  # m3u8 download
 401             },
 402         },
 403         # ooyala video
 404         {
 405             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 406             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 407             'info_dict': {
 408                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 409                 'ext': 'mp4',
 410                 'title': '2cc213299525360.mov',  # that's what we get
 411                 'duration': 238.231,
 412             },
 413             'add_ie': ['Ooyala'],
 414         },
 415         {
 416             # ooyala video embedded with http://player.ooyala.com/iframe.js
 417             'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/',
 418             'info_dict': {
 419                 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB',
 420                 'ext': 'mp4',
 421                 'title': '"Steve Jobs: Man in the Machine" trailer',
 422                 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
 423                 'duration': 135.427,
 424             },
 425             'params': {
 426                 'skip_download': True,
 427             },
 428             'skip': 'movie expired',
 429         },
 430         # embed.ly video
 431         {
 432             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 433             'info_dict': {
 434                 'id': '9ODmcdjQcHQ',
 435                 'ext': 'mp4',
 436                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 437                 'upload_date': '20140225',
 438                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 439                 'uploader': 'Tested',
 440                 'uploader_id': 'testedcom',
 441             },
 442             # No need to test YoutubeIE here
 443             'params': {
 444                 'skip_download': True,
 445             },
 446         },
 447         # funnyordie embed
 448         {
 449             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 450             'info_dict': {
 451                 'id': '18e820ec3f',
 452                 'ext': 'mp4',
 453                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 454                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 455             },
 456             # HEAD requests lead to endless 301, while GET is OK
 457             'expected_warnings': ['301'],
 458         },
 459         # RUTV embed
 460         {
 461             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 462             'info_dict': {
 463                 'id': '776940',
 464                 'ext': 'mp4',
 465                 'title': 'Охотское море стало целиком российским',
 466                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 467             },
 468             'params': {
 469                 # m3u8 download
 470                 'skip_download': True,
 471             },
 472         },
 473         # TVC embed
 474         {
 475             '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/',
 476             'info_dict': {
 477                 'id': '55304',
 478                 'ext': 'mp4',
 479                 'title': 'Дошкольное воспитание',
 480             },
 481         },
 482         # SportBox embed
 483         {
 484             'url': 'http://www.vestifinance.ru/articles/25753',
 485             'info_dict': {
 486                 'id': '25753',
 487                 'title': 'Прямые трансляции с Форума-выставки "Госзаказ-2013"',
 488             },
 489             'playlist': [{
 490                 'info_dict': {
 491                     'id': '370908',
 492                     'title': 'Госзаказ. День 3',
 493                     'ext': 'mp4',
 494                 }
 495             }, {
 496                 'info_dict': {
 497                     'id': '370905',
 498                     'title': 'Госзаказ. День 2',
 499                     'ext': 'mp4',
 500                 }
 501             }, {
 502                 'info_dict': {
 503                     'id': '370902',
 504                     'title': 'Госзаказ. День 1',
 505                     'ext': 'mp4',
 506                 }
 507             }],
 508             'params': {
 509                 # m3u8 download
 510                 'skip_download': True,
 511             },
 512         },
 513         # Myvi.ru embed
 514         {
 515             'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
 516             'info_dict': {
 517                 'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
 518                 'ext': 'mp4',
 519                 'title': 'Ужастики, русский трейлер (2015)',
 520                 'thumbnail': 're:^https?://.*\.jpg$',
 521                 'duration': 153,
 522             }
 523         },
 524         # XHamster embed
 525         {
 526             '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',
 527             'info_dict': {
 528                 'id': 'showthread',
 529                 'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
 530             },
 531             'playlist_mincount': 7,
 532             # This forum does not allow <iframe> syntaxes anymore
 533             # Now HTML tags are displayed as-is
 534             'skip': 'No videos on this page',
 535         },
 536         # Embedded TED video
 537         {
 538             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 539             'md5': '65fdff94098e4a607385a60c5177c638',
 540             'info_dict': {
 541                 'id': '1969',
 542                 'ext': 'mp4',
 543                 'title': 'Hidden miracles of the natural world',
 544                 'uploader': 'Louie Schwartzberg',
 545                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 546             }
 547         },
 548         # Embedded Ustream video
 549         {
 550             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 551             'md5': '27b99cdb639c9b12a79bca876a073417',
 552             'info_dict': {
 553                 'id': '45734260',
 554                 'ext': 'flv',
 555                 'uploader': 'AU SPA:  The NSA and Privacy',
 556                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 557             }
 558         },
 559         # nowvideo embed hidden behind percent encoding
 560         {
 561             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 562             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 563             'info_dict': {
 564                 'id': '06e53103ca9aa',
 565                 'ext': 'flv',
 566                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 567                 'description': 'No description',
 568             },
 569         },
 570         # arte embed
 571         {
 572             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 573             'md5': '7653032cbb25bf6c80d80f217055fa43',
 574             'info_dict': {
 575                 'id': '048195-004_PLUS7-F',
 576                 'ext': 'flv',
 577                 'title': 'X:enius',
 578                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 579                 'upload_date': '20140320',
 580             },
 581             'params': {
 582                 'skip_download': 'Requires rtmpdump'
 583             },
 584             'skip': 'video gone',
 585         },
 586         # francetv embed
 587         {
 588             'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
 589             'info_dict': {
 590                 'id': 'EV_30231',
 591                 'ext': 'mp4',
 592                 'title': 'Alcaline, le concert avec Calogero',
 593                 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
 594                 'upload_date': '20150226',
 595                 'timestamp': 1424989860,
 596                 'duration': 5400,
 597             },
 598             'params': {
 599                 # m3u8 downloads
 600                 'skip_download': True,
 601             },
 602             'expected_warnings': [
 603                 'Forbidden'
 604             ]
 605         },
 606         # Condé Nast embed
 607         {
 608             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 609             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 610             'info_dict': {
 611                 'id': '53501be369702d3275860000',
 612                 'ext': 'mp4',
 613                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 614             }
 615         },
 616         # Dailymotion embed
 617         {
 618             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 619             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 620             'info_dict': {
 621                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 622                 'ext': 'mp4',
 623                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 624                 'description': 'md5:faf028e48a461b8b7fad38f1e104b119',
 625                 'uploader': 'Spi0n',
 626                 'uploader_id': 'xgditw',
 627                 'upload_date': '20140425',
 628                 'timestamp': 1398441542,
 629             },
 630             'add_ie': ['Dailymotion'],
 631         },
 632         # YouTube embed
 633         {
 634             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 635             'info_dict': {
 636                 'id': 'FXRb4ykk4S0',
 637                 'ext': 'mp4',
 638                 'title': 'The NBL Auction 2014',
 639                 'uploader': 'BADMINTON England',
 640                 'uploader_id': 'BADMINTONEvents',
 641                 'upload_date': '20140603',
 642                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 643             },
 644             'add_ie': ['Youtube'],
 645             'params': {
 646                 'skip_download': True,
 647             }
 648         },
 649         # MTVSercices embed
 650         {
 651             'url': 'http://www.vulture.com/2016/06/new-key-peele-sketches-released.html',
 652             'md5': 'ca1aef97695ef2c1d6973256a57e5252',
 653             'info_dict': {
 654                 'id': '769f7ec0-0692-4d62-9b45-0d88074bffc1',
 655                 'ext': 'mp4',
 656                 'title': 'Key and Peele|October 10, 2012|2|203|Liam Neesons - Uncensored',
 657                 'description': 'Two valets share their love for movie star Liam Neesons.',
 658                 'timestamp': 1349922600,
 659                 'upload_date': '20121011',
 660             },
 661         },
 662         # YouTube embed via <data-embed-url="">
 663         {
 664             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 665             'info_dict': {
 666                 'id': '4vAffPZIT44',
 667                 'ext': 'mp4',
 668                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 669                 'uploader': 'Gameloft',
 670                 'uploader_id': 'gameloft',
 671                 'upload_date': '20140828',
 672                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 673             },
 674             'params': {
 675                 'skip_download': True,
 676             }
 677         },
 678         # Camtasia studio
 679         {
 680             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 681             'playlist': [{
 682                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 683                 'info_dict': {
 684                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 685                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 686                     'ext': 'flv',
 687                     'duration': 2235.90,
 688                 }
 689             }, {
 690                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 691                 'info_dict': {
 692                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 693                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 694                     'ext': 'flv',
 695                     'duration': 2235.93,
 696                 }
 697             }],
 698             'info_dict': {
 699                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 700             }
 701         },
 702         # Flowplayer
 703         {
 704             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 705             'md5': '9d65602bf31c6e20014319c7d07fba27',
 706             'info_dict': {
 707                 'id': '5123ea6d5e5a7',
 708                 'ext': 'mp4',
 709                 'age_limit': 18,
 710                 'uploader': 'www.handjobhub.com',
 711                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 712             }
 713         },
 714         # Multiple brightcove videos
 715         # https://github.com/rg3/youtube-dl/issues/2283
 716         {
 717             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 718             'info_dict': {
 719                 'id': 'always-never',
 720                 'title': 'Always / Never - The New Yorker',
 721             },
 722             'playlist_count': 3,
 723             'params': {
 724                 'extract_flat': False,
 725                 'skip_download': True,
 726             }
 727         },
 728         # MLB embed
 729         {
 730             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 731             'md5': '96f09a37e44da40dd083e12d9a683327',
 732             'info_dict': {
 733                 'id': '33322633',
 734                 'ext': 'mp4',
 735                 'title': 'Ump changes call to ball',
 736                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 737                 'duration': 48,
 738                 'timestamp': 1401537900,
 739                 'upload_date': '20140531',
 740                 'thumbnail': 're:^https?://.*\.jpg$',
 741             },
 742         },
 743         # Wistia embed
 744         {
 745             'url': 'http://study.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 746             'md5': '1953f3a698ab51cfc948ed3992a0b7ff',
 747             'info_dict': {
 748                 'id': '6e2wtrbdaf',
 749                 'ext': 'mov',
 750                 'title': 'paywall_north-american-exploration-failed-colonies-of-spain-france-england',
 751                 'description': 'a Paywall Videos video from Remilon',
 752                 'duration': 644.072,
 753                 'uploader': 'study.com',
 754                 'timestamp': 1459678540,
 755                 'upload_date': '20160403',
 756                 'filesize': 24687186,
 757             },
 758         },
 759         {
 760             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 761             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 762             'info_dict': {
 763                 'id': 'uxjb0lwrcz',
 764                 'ext': 'mp4',
 765                 'title': 'Conversation about Hexagonal Rails Part 1',
 766                 'description': 'a Martin Fowler video from ThoughtWorks',
 767                 'duration': 1715.0,
 768                 'uploader': 'thoughtworks.wistia.com',
 769                 'timestamp': 1401832161,
 770                 'upload_date': '20140603',
 771             },
 772         },
 773         # Wistia standard embed (async)
 774         {
 775             'url': 'https://www.getdrip.com/university/brennan-dunn-drip-workshop/',
 776             'info_dict': {
 777                 'id': '807fafadvk',
 778                 'ext': 'mp4',
 779                 'title': 'Drip Brennan Dunn Workshop',
 780                 'description': 'a JV Webinars video from getdrip-1',
 781                 'duration': 4986.95,
 782                 'timestamp': 1463607249,
 783                 'upload_date': '20160518',
 784             },
 785             'params': {
 786                 'skip_download': True,
 787             }
 788         },
 789         # Soundcloud embed
 790         {
 791             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 792             'info_dict': {
 793                 'id': '174391317',
 794                 'ext': 'mp3',
 795                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 796                 'uploader': 'Sophos Security',
 797                 'title': 'Chet Chat 171 - Oct 29, 2014',
 798                 'upload_date': '20141029',
 799             }
 800         },
 801         # Soundcloud multiple embeds
 802         {
 803             'url': 'http://www.guitarplayer.com/lessons/1014/legato-workout-one-hour-to-more-fluid-performance---tab/52809',
 804             'info_dict': {
 805                 'id': '52809',
 806                 'title': 'Guitar Essentials: Legato Workout—One-Hour to Fluid Performance  | TAB + AUDIO',
 807             },
 808             'playlist_mincount': 7,
 809         },
 810         # Livestream embed
 811         {
 812             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 813             'info_dict': {
 814                 'id': '67864563',
 815                 'ext': 'flv',
 816                 'upload_date': '20141112',
 817                 'title': 'Rosetta #CometLanding webcast HL 10',
 818             }
 819         },
 820         # Another Livestream embed, without 'new.' in URL
 821         {
 822             'url': 'https://www.freespeech.org/',
 823             'info_dict': {
 824                 'id': '123537347',
 825                 'ext': 'mp4',
 826                 'title': 're:^FSTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 827             },
 828             'params': {
 829                 # Live stream
 830                 'skip_download': True,
 831             },
 832         },
 833         # LazyYT
 834         {
 835             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
 836             'info_dict': {
 837                 'id': '1986',
 838                 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
 839             },
 840             'playlist_mincount': 2,
 841         },
 842         # Cinchcast embed
 843         {
 844             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
 845             'info_dict': {
 846                 'id': '7141703',
 847                 'ext': 'mp3',
 848                 'upload_date': '20141126',
 849                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
 850             }
 851         },
 852         # Cinerama player
 853         {
 854             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
 855             'info_dict': {
 856                 'id': '730m_DandD_1901_512k',
 857                 'ext': 'mp4',
 858                 'uploader': 'www.abc.net.au',
 859                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
 860             }
 861         },
 862         # embedded viddler video
 863         {
 864             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
 865             'info_dict': {
 866                 'id': '4d03aad9',
 867                 'ext': 'mp4',
 868                 'uploader': 'deadspin',
 869                 'title': 'WALL-TO-GORTAT',
 870                 'timestamp': 1422285291,
 871                 'upload_date': '20150126',
 872             },
 873             'add_ie': ['Viddler'],
 874         },
 875         # Libsyn embed
 876         {
 877             'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
 878             'info_dict': {
 879                 'id': '3377616',
 880                 'ext': 'mp3',
 881                 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
 882                 'description': 'md5:601cb790edd05908957dae8aaa866465',
 883                 'upload_date': '20150220',
 884             },
 885             'skip': 'All The Daily Show URLs now redirect to http://www.cc.com/shows/',
 886         },
 887         # jwplayer YouTube
 888         {
 889             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
 890             'info_dict': {
 891                 'id': 'Mrj4DVp2zeA',
 892                 'ext': 'mp4',
 893                 'upload_date': '20150212',
 894                 'uploader': 'The National Archives UK',
 895                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
 896                 'uploader_id': 'NationalArchives08',
 897                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
 898             },
 899         },
 900         # rtl.nl embed
 901         {
 902             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
 903             'playlist_mincount': 5,
 904             'info_dict': {
 905                 'id': 'aanslagen-kopenhagen',
 906                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
 907             }
 908         },
 909         # Zapiks embed
 910         {
 911             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
 912             'info_dict': {
 913                 'id': '118046',
 914                 'ext': 'mp4',
 915                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
 916             }
 917         },
 918         # Kaltura embed (different embed code)
 919         {
 920             'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
 921             'info_dict': {
 922                 'id': '1_a52wc67y',
 923                 'ext': 'flv',
 924                 'upload_date': '20150127',
 925                 'uploader_id': 'PremierMedia',
 926                 'timestamp': int,
 927                 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
 928             },
 929         },
 930         # Kaltura embed protected with referrer
 931         {
 932             'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero',
 933             'info_dict': {
 934                 'id': '1_g4fbemnq',
 935                 'ext': 'mp4',
 936                 'title': 'Violetta - Achter De Schermen - Ruggero',
 937                 'description': 'Achter de schermen met Ruggero',
 938                 'timestamp': 1435133761,
 939                 'upload_date': '20150624',
 940                 'uploader_id': 'echojecka',
 941             },
 942         },
 943         # Kaltura embed with single quotes
 944         {
 945             'url': 'http://fod.infobase.com/p_ViewPlaylist.aspx?AssignmentID=NUN8ZY',
 946             'info_dict': {
 947                 'id': '0_izeg5utt',
 948                 'ext': 'mp4',
 949                 'title': '35871',
 950                 'timestamp': 1355743100,
 951                 'upload_date': '20121217',
 952                 'uploader_id': 'batchUser',
 953             },
 954             'add_ie': ['Kaltura'],
 955         },
 956         {
 957             # Kaltura embedded via quoted entry_id
 958             'url': 'https://www.oreilly.com/ideas/my-cloud-makes-pretty-pictures',
 959             'info_dict': {
 960                 'id': '0_utuok90b',
 961                 'ext': 'mp4',
 962                 'title': '06_matthew_brender_raj_dutt',
 963                 'timestamp': 1466638791,
 964                 'upload_date': '20160622',
 965             },
 966             'add_ie': ['Kaltura'],
 967             'expected_warnings': [
 968                 'Could not send HEAD request'
 969             ],
 970             'params': {
 971                 'skip_download': True,
 972             }
 973         },
 974         # Eagle.Platform embed (generic URL)
 975         {
 976             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
 977             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
 978             'info_dict': {
 979                 'id': '227304',
 980                 'ext': 'mp4',
 981                 'title': 'Навальный вышел на свободу',
 982                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
 983                 'thumbnail': 're:^https?://.*\.jpg$',
 984                 'duration': 87,
 985                 'view_count': int,
 986                 'age_limit': 0,
 987             },
 988         },
 989         # ClipYou (Eagle.Platform) embed (custom URL)
 990         {
 991             'url': 'http://muz-tv.ru/play/7129/',
 992             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
 993             'info_dict': {
 994                 'id': '12820',
 995                 'ext': 'mp4',
 996                 'title': "'O Sole Mio",
 997                 'thumbnail': 're:^https?://.*\.jpg$',
 998                 'duration': 216,
 999                 'view_count': int,
1000             },
1001         },
1002         # Pladform embed
1003         {
1004             'url': 'http://muz-tv.ru/kinozal/view/7400/',
1005             'info_dict': {
1006                 'id': '100183293',
1007                 'ext': 'mp4',
1008                 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
1009                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
1010                 'thumbnail': 're:^https?://.*\.jpg$',
1011                 'duration': 694,
1012                 'age_limit': 0,
1013             },
1014         },
1015         # Playwire embed
1016         {
1017             'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
1018             'info_dict': {
1019                 'id': '3519514',
1020                 'ext': 'mp4',
1021                 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
1022                 'thumbnail': 're:^https?://.*\.png$',
1023                 'duration': 45.115,
1024             },
1025         },
1026         # 5min embed
1027         {
1028             'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
1029             'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
1030             'info_dict': {
1031                 'id': '518726732',
1032                 'ext': 'mp4',
1033                 'title': 'Facebook Creates "On This Day" | Crunch Report',
1034             },
1035         },
1036         # SVT embed
1037         {
1038             'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
1039             'info_dict': {
1040                 'id': '2900353',
1041                 'ext': 'flv',
1042                 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
1043                 'duration': 27,
1044                 'age_limit': 0,
1045             },
1046         },
1047         # Crooks and Liars embed
1048         {
1049             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
1050             'info_dict': {
1051                 'id': '8RUoRhRi',
1052                 'ext': 'mp4',
1053                 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
1054                 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
1055                 'timestamp': 1428207000,
1056                 'upload_date': '20150405',
1057                 'uploader': 'Heather',
1058             },
1059         },
1060         # Crooks and Liars external embed
1061         {
1062             'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
1063             'info_dict': {
1064                 'id': 'MTE3MjUtMzQ2MzA',
1065                 'ext': 'mp4',
1066                 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
1067                 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
1068                 'timestamp': 1265032391,
1069                 'upload_date': '20100201',
1070                 'uploader': 'Heather',
1071             },
1072         },
1073         # NBC Sports vplayer embed
1074         {
1075             'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
1076             'info_dict': {
1077                 'id': 'ln7x1qSThw4k',
1078                 'ext': 'flv',
1079                 'title': "PFT Live: New leader in the 'new-look' defense",
1080                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
1081                 'uploader': 'NBCU-SPORTS',
1082                 'upload_date': '20140107',
1083                 'timestamp': 1389118457,
1084             },
1085         },
1086         # NBC News embed
1087         {
1088             'url': 'http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html',
1089             'md5': '1aa589c675898ae6d37a17913cf68d66',
1090             'info_dict': {
1091                 'id': '701714499682',
1092                 'ext': 'mp4',
1093                 'title': 'PREVIEW: On Assignment: David Letterman',
1094                 '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.',
1095             },
1096         },
1097         # UDN embed
1098         {
1099             'url': 'https://video.udn.com/news/300346',
1100             'md5': 'fd2060e988c326991037b9aff9df21a6',
1101             'info_dict': {
1102                 'id': '300346',
1103                 'ext': 'mp4',
1104                 'title': '中一中男師變性 全校師生力挺',
1105                 'thumbnail': 're:^https?://.*\.jpg$',
1106             },
1107             'params': {
1108                 # m3u8 download
1109                 'skip_download': True,
1110             },
1111         },
1112         # Ooyala embed
1113         {
1114             'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
1115             'info_dict': {
1116                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
1117                 'ext': 'mp4',
1118                 'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.',
1119                 'title': 'This is what separates the Excel masters from the wannabes',
1120                 'duration': 191.933,
1121             },
1122             'params': {
1123                 # m3u8 downloads
1124                 'skip_download': True,
1125             }
1126         },
1127         # Brightcove URL in single quotes
1128         {
1129             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
1130             'md5': '4ae374f1f8b91c889c4b9203c8c752af',
1131             'info_dict': {
1132                 'id': '4255764656001',
1133                 'ext': 'mp4',
1134                 'title': 'SN Presents: Russell Martin, World Citizen',
1135                 '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.',
1136                 'uploader': 'Rogers Sportsnet',
1137                 'uploader_id': '1704050871',
1138                 'upload_date': '20150525',
1139                 'timestamp': 1432570283,
1140             },
1141         },
1142         # Dailymotion Cloud video
1143         {
1144             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
1145             'md5': 'dcaf23ad0c67a256f4278bce6e0bae38',
1146             'info_dict': {
1147                 'id': 'x2uy8t3',
1148                 'ext': 'mp4',
1149                 'title': 'Sauvons les abeilles ! - Le débat',
1150                 'description': 'md5:d9082128b1c5277987825d684939ca26',
1151                 'thumbnail': 're:^https?://.*\.jpe?g$',
1152                 'timestamp': 1434970506,
1153                 'upload_date': '20150622',
1154                 'uploader': 'Public Sénat',
1155                 'uploader_id': 'xa9gza',
1156             }
1157         },
1158         # OnionStudios embed
1159         {
1160             'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
1161             'info_dict': {
1162                 'id': '2855',
1163                 'ext': 'mp4',
1164                 'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
1165                 'thumbnail': 're:^https?://.*\.jpe?g$',
1166                 'uploader': 'ClickHole',
1167                 'uploader_id': 'clickhole',
1168             }
1169         },
1170         # SnagFilms embed
1171         {
1172             'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
1173             'info_dict': {
1174                 'id': '74849a00-85a9-11e1-9660-123139220831',
1175                 'ext': 'mp4',
1176                 'title': '#whilewewatch',
1177             }
1178         },
1179         # AdobeTVVideo embed
1180         {
1181             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
1182             'md5': '43662b577c018ad707a63766462b1e87',
1183             'info_dict': {
1184                 'id': '2456',
1185                 'ext': 'mp4',
1186                 'title': 'New experience with Acrobat DC',
1187                 'description': 'New experience with Acrobat DC',
1188                 'duration': 248.667,
1189             },
1190         },
1191         # ScreenwaveMedia embed
1192         {
1193             'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1',
1194             'md5': '24ace5baba0d35d55c6810b51f34e9e0',
1195             'info_dict': {
1196                 'id': 'cinemasnob-55d26273809dd',
1197                 'ext': 'mp4',
1198                 'title': 'cinemasnob',
1199             },
1200         },
1201         # BrightcoveInPageEmbed embed
1202         {
1203             'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
1204             'info_dict': {
1205                 'id': '4238694884001',
1206                 'ext': 'flv',
1207                 'title': 'Tabletop: Dread, Last Thoughts',
1208                 'description': 'Tabletop: Dread, Last Thoughts',
1209                 'duration': 51690,
1210             },
1211         },
1212         # JWPlayer with M3U8
1213         {
1214             'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video',
1215             'info_dict': {
1216                 'id': 'playlist',
1217                 'ext': 'mp4',
1218                 'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ',
1219                 'uploader': 'ren.tv',
1220             },
1221             'params': {
1222                 # m3u8 downloads
1223                 'skip_download': True,
1224             }
1225         },
1226         # Brightcove embed, with no valid 'renditions' but valid 'IOSRenditions'
1227         # This video can't be played in browsers if Flash disabled and UA set to iPhone, which is actually a false alarm
1228         {
1229             'url': 'https://dl.dropboxusercontent.com/u/29092637/interview.html',
1230             'info_dict': {
1231                 'id': '4785848093001',
1232                 'ext': 'mp4',
1233                 'title': 'The Cardinal Pell Interview',
1234                 'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ',
1235                 'uploader': 'GlobeCast Australia - GlobeStream',
1236                 'uploader_id': '2733773828001',
1237                 'upload_date': '20160304',
1238                 'timestamp': 1457083087,
1239             },
1240             'params': {
1241                 # m3u8 downloads
1242                 'skip_download': True,
1243             },
1244         },
1245         # Another form of arte.tv embed
1246         {
1247             'url': 'http://www.tv-replay.fr/redirection/09-04-16/arte-reportage-arte-11508975.html',
1248             'md5': '850bfe45417ddf221288c88a0cffe2e2',
1249             'info_dict': {
1250                 'id': '030273-562_PLUS7-F',
1251                 'ext': 'mp4',
1252                 'title': 'ARTE Reportage - Nulle part, en France',
1253                 'description': 'md5:e3a0e8868ed7303ed509b9e3af2b870d',
1254                 'upload_date': '20160409',
1255             },
1256         },
1257         # LiveLeak embed
1258         {
1259             'url': 'http://www.wykop.pl/link/3088787/',
1260             'md5': 'ace83b9ed19b21f68e1b50e844fdf95d',
1261             'info_dict': {
1262                 'id': '874_1459135191',
1263                 'ext': 'mp4',
1264                 'title': 'Man shows poor quality of new apartment building',
1265                 'description': 'The wall is like a sand pile.',
1266                 'uploader': 'Lake8737',
1267             }
1268         },
1269         # Duplicated embedded video URLs
1270         {
1271             'url': 'http://www.hudl.com/athlete/2538180/highlights/149298443',
1272             'info_dict': {
1273                 'id': '149298443_480_16c25b74_2',
1274                 'ext': 'mp4',
1275                 'title': 'vs. Blue Orange Spring Game',
1276                 'uploader': 'www.hudl.com',
1277             },
1278         },
1279         # twitter:player:stream embed
1280         {
1281             'url': 'http://www.rtl.be/info/video/589263.aspx?CategoryID=288',
1282             'info_dict': {
1283                 'id': 'master',
1284                 'ext': 'mp4',
1285                 'title': 'Une nouvelle espèce de dinosaure découverte en Argentine',
1286                 'uploader': 'www.rtl.be',
1287             },
1288             'params': {
1289                 # m3u8 downloads
1290                 'skip_download': True,
1291             },
1292         },
1293         # twitter:player embed
1294         {
1295             'url': 'http://www.theatlantic.com/video/index/484130/what-do-black-holes-sound-like/',
1296             'md5': 'a3e0df96369831de324f0778e126653c',
1297             'info_dict': {
1298                 'id': '4909620399001',
1299                 'ext': 'mp4',
1300                 'title': 'What Do Black Holes Sound Like?',
1301                 'description': 'what do black holes sound like',
1302                 'upload_date': '20160524',
1303                 'uploader_id': '29913724001',
1304                 'timestamp': 1464107587,
1305                 'uploader': 'TheAtlantic',
1306             },
1307             'add_ie': ['BrightcoveLegacy'],
1308         },
1309         # Facebook <iframe> embed
1310         {
1311             'url': 'https://www.hostblogger.de/blog/archives/6181-Auto-jagt-Betonmischer.html',
1312             'md5': 'fbcde74f534176ecb015849146dd3aee',
1313             'info_dict': {
1314                 'id': '599637780109885',
1315                 'ext': 'mp4',
1316                 'title': 'Facebook video #599637780109885',
1317             },
1318         },
1319         # Facebook API embed
1320         {
1321             'url': 'http://www.lothype.com/blue-stars-2016-preview-standstill-full-show/',
1322             'md5': 'a47372ee61b39a7b90287094d447d94e',
1323             'info_dict': {
1324                 'id': '10153467542406923',
1325                 'ext': 'mp4',
1326                 'title': 'Facebook video #10153467542406923',
1327             },
1328         },
1329         # Wordpress "YouTube Video Importer" plugin
1330         {
1331             'url': 'http://www.lothype.com/blue-devils-drumline-stanford-lot-2016/',
1332             'md5': 'd16797741b560b485194eddda8121b48',
1333             'info_dict': {
1334                 'id': 'HNTXWDXV9Is',
1335                 'ext': 'mp4',
1336                 'title': 'Blue Devils Drumline Stanford lot 2016',
1337                 'upload_date': '20160627',
1338                 'uploader_id': 'GENOCIDE8GENERAL10',
1339                 'uploader': 'cylus cyrus',
1340             },
1341         },
1342         {
1343             # video stored on custom kaltura server
1344             'url': 'http://www.expansion.com/multimedia/videos.html?media=EQcM30NHIPv',
1345             'md5': '537617d06e64dfed891fa1593c4b30cc',
1346             'info_dict': {
1347                 'id': '0_1iotm5bh',
1348                 'ext': 'mp4',
1349                 'title': 'Elecciones británicas: 5 lecciones para Rajoy',
1350                 'description': 'md5:435a89d68b9760b92ce67ed227055f16',
1351                 'uploader_id': 'videos.expansion@el-mundo.net',
1352                 'upload_date': '20150429',
1353                 'timestamp': 1430303472,
1354             },
1355             'add_ie': ['Kaltura'],
1356         },
1357         {
1358             # Non-standard Vimeo embed
1359             'url': 'https://openclassrooms.com/courses/understanding-the-web',
1360             'md5': '64d86f1c7d369afd9a78b38cbb88d80a',
1361             'info_dict': {
1362                 'id': '148867247',
1363                 'ext': 'mp4',
1364                 'title': 'Understanding the web - Teaser',
1365                 'description': 'This is "Understanding the web - Teaser" by openclassrooms on Vimeo, the home for high quality videos and the people who love them.',
1366                 'upload_date': '20151214',
1367                 'uploader': 'OpenClassrooms',
1368                 'uploader_id': 'openclassrooms',
1369             },
1370             'add_ie': ['Vimeo'],
1371         },
1372         {
1373             # generic vimeo embed that requires original URL passed as Referer
1374             'url': 'http://racing4everyone.eu/2016/07/30/formula-1-2016-round12-germany/',
1375             'only_matching': True,
1376         },
1377         {
1378             'url': 'https://support.arkena.com/display/PLAY/Ways+to+embed+your+video',
1379             'md5': 'b96f2f71b359a8ecd05ce4e1daa72365',
1380             'info_dict': {
1381                 'id': 'b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe',
1382                 'ext': 'mp4',
1383                 'title': 'Big Buck Bunny',
1384                 'description': 'Royalty free test video',
1385                 'timestamp': 1432816365,
1386                 'upload_date': '20150528',
1387                 'is_live': False,
1388             },
1389             'params': {
1390                 'skip_download': True,
1391             },
1392             'add_ie': [ArkenaIE.ie_key()],
1393         },
1394         {
1395             '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/',
1396             'info_dict': {
1397                 'id': '1c7141f46c',
1398                 'ext': 'mp4',
1399                 'title': 'НА КОСЪМ ОТ ВЗРИВ: Изтичане на газ на бензиностанция в Пловдив',
1400             },
1401             'params': {
1402                 'skip_download': True,
1403             },
1404             'add_ie': [Vbox7IE.ie_key()],
1405         },
1406         {
1407             # DBTV embeds
1408             'url': 'http://www.dagbladet.no/2016/02/23/nyheter/nordlys/ski/troms/ver/43254897/',
1409             'info_dict': {
1410                 'id': '43254897',
1411                 'title': 'Etter ett års planlegging, klaffet endelig alt: - Jeg måtte ta en liten dans',
1412             },
1413             'playlist_mincount': 3,
1414         },
1415         # {
1416         #     # TODO: find another test
1417         #     # http://schema.org/VideoObject
1418         #     'url': 'https://flipagram.com/f/nyvTSJMKId',
1419         #     'md5': '888dcf08b7ea671381f00fab74692755',
1420         #     'info_dict': {
1421         #         'id': 'nyvTSJMKId',
1422         #         'ext': 'mp4',
1423         #         'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction',
1424         #         'description': '#love for cats.',
1425         #         'timestamp': 1461244995,
1426         #         'upload_date': '20160421',
1427         #     },
1428         #     'params': {
1429         #         'force_generic_extractor': True,
1430         #     },
1431         # }
1432     ]
1433
1434     def report_following_redirect(self, new_url):
1435         """Report information extraction."""
1436         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
1437
1438     def _extract_rss(self, url, video_id, doc):
1439         playlist_title = doc.find('./channel/title').text
1440         playlist_desc_el = doc.find('./channel/description')
1441         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
1442
1443         entries = []
1444         for it in doc.findall('./channel/item'):
1445             next_url = xpath_text(it, 'link', fatal=False)
1446             if not next_url:
1447                 enclosure_nodes = it.findall('./enclosure')
1448                 for e in enclosure_nodes:
1449                     next_url = e.attrib.get('url')
1450                     if next_url:
1451                         break
1452
1453             if not next_url:
1454                 continue
1455
1456             entries.append({
1457                 '_type': 'url',
1458                 'url': next_url,
1459                 'title': it.find('title').text,
1460             })
1461
1462         return {
1463             '_type': 'playlist',
1464             'id': url,
1465             'title': playlist_title,
1466             'description': playlist_desc,
1467             'entries': entries,
1468         }
1469
1470     def _extract_camtasia(self, url, video_id, webpage):
1471         """ Returns None if no camtasia video can be found. """
1472
1473         camtasia_cfg = self._search_regex(
1474             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
1475             webpage, 'camtasia configuration file', default=None)
1476         if camtasia_cfg is None:
1477             return None
1478
1479         title = self._html_search_meta('DC.title', webpage, fatal=True)
1480
1481         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
1482         camtasia_cfg = self._download_xml(
1483             camtasia_url, video_id,
1484             note='Downloading camtasia configuration',
1485             errnote='Failed to download camtasia configuration')
1486         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
1487
1488         entries = []
1489         for n in fileset_node.getchildren():
1490             url_n = n.find('./uri')
1491             if url_n is None:
1492                 continue
1493
1494             entries.append({
1495                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
1496                 'title': '%s - %s' % (title, n.tag),
1497                 'url': compat_urlparse.urljoin(url, url_n.text),
1498                 'duration': float_or_none(n.find('./duration').text),
1499             })
1500
1501         return {
1502             '_type': 'playlist',
1503             'entries': entries,
1504             'title': title,
1505         }
1506
1507     def _real_extract(self, url):
1508         if url.startswith('//'):
1509             return {
1510                 '_type': 'url',
1511                 'url': self.http_scheme() + url,
1512             }
1513
1514         parsed_url = compat_urlparse.urlparse(url)
1515         if not parsed_url.scheme:
1516             default_search = self._downloader.params.get('default_search')
1517             if default_search is None:
1518                 default_search = 'fixup_error'
1519
1520             if default_search in ('auto', 'auto_warning', 'fixup_error'):
1521                 if '/' in url:
1522                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
1523                     return self.url_result('http://' + url)
1524                 elif default_search != 'fixup_error':
1525                     if default_search == 'auto_warning':
1526                         if re.match(r'^(?:url|URL)$', url):
1527                             raise ExtractorError(
1528                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
1529                                 expected=True)
1530                         else:
1531                             self._downloader.report_warning(
1532                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
1533                     return self.url_result('ytsearch:' + url)
1534
1535             if default_search in ('error', 'fixup_error'):
1536                 raise ExtractorError(
1537                     '%r is not a valid URL. '
1538                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
1539                     % (url, url), expected=True)
1540             else:
1541                 if ':' not in default_search:
1542                     default_search += ':'
1543                 return self.url_result(default_search + url)
1544
1545         url, smuggled_data = unsmuggle_url(url)
1546         force_videoid = None
1547         is_intentional = smuggled_data and smuggled_data.get('to_generic')
1548         if smuggled_data and 'force_videoid' in smuggled_data:
1549             force_videoid = smuggled_data['force_videoid']
1550             video_id = force_videoid
1551         else:
1552             video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
1553
1554         self.to_screen('%s: Requesting header' % video_id)
1555
1556         head_req = HEADRequest(url)
1557         head_response = self._request_webpage(
1558             head_req, video_id,
1559             note=False, errnote='Could not send HEAD request to %s' % url,
1560             fatal=False)
1561
1562         if head_response is not False:
1563             # Check for redirect
1564             new_url = head_response.geturl()
1565             if url != new_url:
1566                 self.report_following_redirect(new_url)
1567                 if force_videoid:
1568                     new_url = smuggle_url(
1569                         new_url, {'force_videoid': force_videoid})
1570                 return self.url_result(new_url)
1571
1572         full_response = None
1573         if head_response is False:
1574             request = sanitized_Request(url)
1575             request.add_header('Accept-Encoding', '*')
1576             full_response = self._request_webpage(request, video_id)
1577             head_response = full_response
1578
1579         info_dict = {
1580             'id': video_id,
1581             'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1582             'upload_date': unified_strdate(head_response.headers.get('Last-Modified'))
1583         }
1584
1585         # Check for direct link to a video
1586         content_type = head_response.headers.get('Content-Type', '').lower()
1587         m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>[^;\s]+)', content_type)
1588         if m:
1589             format_id = m.group('format_id')
1590             if format_id.endswith('mpegurl'):
1591                 formats = self._extract_m3u8_formats(url, video_id, 'mp4')
1592             elif format_id == 'f4m':
1593                 formats = self._extract_f4m_formats(url, video_id)
1594             else:
1595                 formats = [{
1596                     'format_id': m.group('format_id'),
1597                     'url': url,
1598                     'vcodec': 'none' if m.group('type') == 'audio' else None
1599                 }]
1600                 info_dict['direct'] = True
1601             self._sort_formats(formats)
1602             info_dict['formats'] = formats
1603             return info_dict
1604
1605         if not self._downloader.params.get('test', False) and not is_intentional:
1606             force = self._downloader.params.get('force_generic_extractor', False)
1607             self._downloader.report_warning(
1608                 '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
1609
1610         if not full_response:
1611             request = sanitized_Request(url)
1612             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
1613             # making it impossible to download only chunk of the file (yet we need only 512kB to
1614             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
1615             # that will always result in downloading the whole file that is not desirable.
1616             # Therefore for extraction pass we have to override Accept-Encoding to any in order
1617             # to accept raw bytes and being able to download only a chunk.
1618             # It may probably better to solve this by checking Content-Type for application/octet-stream
1619             # after HEAD request finishes, but not sure if we can rely on this.
1620             request.add_header('Accept-Encoding', '*')
1621             full_response = self._request_webpage(request, video_id)
1622
1623         first_bytes = full_response.read(512)
1624
1625         # Is it an M3U playlist?
1626         if first_bytes.startswith(b'#EXTM3U'):
1627             info_dict['formats'] = self._extract_m3u8_formats(url, video_id, 'mp4')
1628             self._sort_formats(info_dict['formats'])
1629             return info_dict
1630
1631         # Maybe it's a direct link to a video?
1632         # Be careful not to download the whole thing!
1633         if not is_html(first_bytes):
1634             self._downloader.report_warning(
1635                 'URL could be a direct video link, returning it as such.')
1636             info_dict.update({
1637                 'direct': True,
1638                 'url': url,
1639             })
1640             return info_dict
1641
1642         webpage = self._webpage_read_content(
1643             full_response, url, video_id, prefix=first_bytes)
1644
1645         self.report_extraction(video_id)
1646
1647         # Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest?
1648         try:
1649             doc = compat_etree_fromstring(webpage.encode('utf-8'))
1650             if doc.tag == 'rss':
1651                 return self._extract_rss(url, video_id, doc)
1652             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
1653                 smil = self._parse_smil(doc, url, video_id)
1654                 self._sort_formats(smil['formats'])
1655                 return smil
1656             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
1657                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
1658             elif re.match(r'(?i)^(?:{[^}]+})?MPD$', doc.tag):
1659                 info_dict['formats'] = self._parse_mpd_formats(
1660                     doc, video_id, mpd_base_url=url.rpartition('/')[0])
1661                 self._sort_formats(info_dict['formats'])
1662                 return info_dict
1663             elif re.match(r'^{http://ns\.adobe\.com/f4m/[12]\.0}manifest$', doc.tag):
1664                 info_dict['formats'] = self._parse_f4m_formats(doc, url, video_id)
1665                 self._sort_formats(info_dict['formats'])
1666                 return info_dict
1667         except compat_xml_parse_error:
1668             pass
1669
1670         # Is it a Camtasia project?
1671         camtasia_res = self._extract_camtasia(url, video_id, webpage)
1672         if camtasia_res is not None:
1673             return camtasia_res
1674
1675         # Sometimes embedded video player is hidden behind percent encoding
1676         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1677         # Unescaping the whole page allows to handle those cases in a generic way
1678         webpage = compat_urllib_parse_unquote(webpage)
1679
1680         # it's tempting to parse this further, but you would
1681         # have to take into account all the variations like
1682         #   Video Title - Site Name
1683         #   Site Name | Video Title
1684         #   Video Title - Tagline | Site Name
1685         # and so on and so forth; it's just not practical
1686         video_title = self._og_search_title(
1687             webpage, default=None) or self._html_search_regex(
1688             r'(?s)<title>(.*?)</title>', webpage, 'video title',
1689             default='video')
1690
1691         # Try to detect age limit automatically
1692         age_limit = self._rta_search(webpage)
1693         # And then there are the jokers who advertise that they use RTA,
1694         # but actually don't.
1695         AGE_LIMIT_MARKERS = [
1696             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1697         ]
1698         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1699             age_limit = 18
1700
1701         # video uploader is domain name
1702         video_uploader = self._search_regex(
1703             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1704
1705         video_description = self._og_search_description(webpage, default=None)
1706         video_thumbnail = self._og_search_thumbnail(webpage, default=None)
1707
1708         # Helper method
1709         def _playlist_from_matches(matches, getter=None, ie=None):
1710             urlrs = orderedSet(
1711                 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1712                 for m in matches)
1713             return self.playlist_result(
1714                 urlrs, playlist_id=video_id, playlist_title=video_title)
1715
1716         # Look for Brightcove Legacy Studio embeds
1717         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
1718         if bc_urls:
1719             self.to_screen('Brightcove video detected.')
1720             entries = [{
1721                 '_type': 'url',
1722                 'url': smuggle_url(bc_url, {'Referer': url}),
1723                 'ie_key': 'BrightcoveLegacy'
1724             } for bc_url in bc_urls]
1725
1726             return {
1727                 '_type': 'playlist',
1728                 'title': video_title,
1729                 'id': video_id,
1730                 'entries': entries,
1731             }
1732
1733         # Look for Brightcove New Studio embeds
1734         bc_urls = BrightcoveNewIE._extract_urls(webpage)
1735         if bc_urls:
1736             return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
1737
1738         # Look for ThePlatform embeds
1739         tp_urls = ThePlatformIE._extract_urls(webpage)
1740         if tp_urls:
1741             return _playlist_from_matches(tp_urls, ie='ThePlatform')
1742
1743         # Look for Vessel embeds
1744         vessel_urls = VesselIE._extract_urls(webpage)
1745         if vessel_urls:
1746             return _playlist_from_matches(vessel_urls, ie=VesselIE.ie_key())
1747
1748         # Look for embedded rtl.nl player
1749         matches = re.findall(
1750             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1751             webpage)
1752         if matches:
1753             return _playlist_from_matches(matches, ie='RtlNl')
1754
1755         vimeo_url = VimeoIE._extract_vimeo_url(url, webpage)
1756         if vimeo_url is not None:
1757             return self.url_result(vimeo_url)
1758
1759         vid_me_embed_url = self._search_regex(
1760             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
1761             webpage, 'vid.me embed', default=None)
1762         if vid_me_embed_url is not None:
1763             return self.url_result(vid_me_embed_url, 'Vidme')
1764
1765         # Look for embedded YouTube player
1766         matches = re.findall(r'''(?x)
1767             (?:
1768                 <iframe[^>]+?src=|
1769                 data-video-url=|
1770                 <embed[^>]+?src=|
1771                 embedSWF\(?:\s*|
1772                 new\s+SWFObject\(
1773             )
1774             (["\'])
1775                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1776                 (?:embed|v|p)/.+?)
1777             \1''', webpage)
1778         if matches:
1779             return _playlist_from_matches(
1780                 matches, lambda m: unescapeHTML(m[1]))
1781
1782         # Look for lazyYT YouTube embed
1783         matches = re.findall(
1784             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1785         if matches:
1786             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1787
1788         # Look for Wordpress "YouTube Video Importer" plugin
1789         matches = re.findall(r'''(?x)<div[^>]+
1790             class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1791             data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1792         if matches:
1793             return _playlist_from_matches(matches, lambda m: m[-1])
1794
1795         matches = DailymotionIE._extract_urls(webpage)
1796         if matches:
1797             return _playlist_from_matches(matches)
1798
1799         # Look for embedded Dailymotion playlist player (#3822)
1800         m = re.search(
1801             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1802         if m:
1803             playlists = re.findall(
1804                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1805             if playlists:
1806                 return _playlist_from_matches(
1807                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1808
1809         # Look for embedded Wistia player
1810         match = re.search(
1811             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1812         if match:
1813             embed_url = self._proto_relative_url(
1814                 unescapeHTML(match.group('url')))
1815             return {
1816                 '_type': 'url_transparent',
1817                 'url': embed_url,
1818                 'ie_key': 'Wistia',
1819                 'uploader': video_uploader,
1820             }
1821
1822         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1823         if match:
1824             return {
1825                 '_type': 'url_transparent',
1826                 'url': 'wistia:%s' % match.group('id'),
1827                 'ie_key': 'Wistia',
1828                 'uploader': video_uploader,
1829             }
1830
1831         match = re.search(
1832             r'''(?sx)
1833                 <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*?
1834                 <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]+)\b.*?\2
1835             ''', webpage)
1836         if match:
1837             return self.url_result(self._proto_relative_url(
1838                 'wistia:%s' % match.group('id')), 'Wistia')
1839
1840         # Look for SVT player
1841         svt_url = SVTIE._extract_url(webpage)
1842         if svt_url:
1843             return self.url_result(svt_url, 'SVT')
1844
1845         # Look for embedded condenast player
1846         matches = re.findall(
1847             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1848             webpage)
1849         if matches:
1850             return {
1851                 '_type': 'playlist',
1852                 'entries': [{
1853                     '_type': 'url',
1854                     'ie_key': 'CondeNast',
1855                     'url': ma,
1856                 } for ma in matches],
1857                 'title': video_title,
1858                 'id': video_id,
1859             }
1860
1861         # Look for Bandcamp pages with custom domain
1862         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1863         if mobj is not None:
1864             burl = unescapeHTML(mobj.group(1))
1865             # Don't set the extractor because it can be a track url or an album
1866             return self.url_result(burl)
1867
1868         # Look for embedded Vevo player
1869         mobj = re.search(
1870             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1871         if mobj is not None:
1872             return self.url_result(mobj.group('url'))
1873
1874         # Look for embedded Viddler player
1875         mobj = re.search(
1876             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1877             webpage)
1878         if mobj is not None:
1879             return self.url_result(mobj.group('url'))
1880
1881         # Look for NYTimes player
1882         mobj = re.search(
1883             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1884             webpage)
1885         if mobj is not None:
1886             return self.url_result(mobj.group('url'))
1887
1888         # Look for Libsyn player
1889         mobj = re.search(
1890             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1891         if mobj is not None:
1892             return self.url_result(mobj.group('url'))
1893
1894         # Look for Ooyala videos
1895         mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1896                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1897                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1898                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1899         if mobj is not None:
1900             return OoyalaIE._build_url_result(smuggle_url(mobj.group('ec'), {'domain': url}))
1901
1902         # Look for multiple Ooyala embeds on SBN network websites
1903         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1904         if mobj is not None:
1905             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1906             if embeds:
1907                 return _playlist_from_matches(
1908                     embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
1909
1910         # Look for Aparat videos
1911         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1912         if mobj is not None:
1913             return self.url_result(mobj.group(1), 'Aparat')
1914
1915         # Look for MPORA videos
1916         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1917         if mobj is not None:
1918             return self.url_result(mobj.group(1), 'Mpora')
1919
1920         # Look for embedded NovaMov-based player
1921         mobj = re.search(
1922             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1923                     (?P<url>http://(?:(?:embed|www)\.)?
1924                         (?:novamov\.com|
1925                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
1926                            videoweed\.(?:es|com)|
1927                            movshare\.(?:net|sx|ag)|
1928                            divxstage\.(?:eu|net|ch|co|at|ag))
1929                         /embed\.php.+?)\1''', webpage)
1930         if mobj is not None:
1931             return self.url_result(mobj.group('url'))
1932
1933         # Look for embedded Facebook player
1934         facebook_url = FacebookIE._extract_url(webpage)
1935         if facebook_url is not None:
1936             return self.url_result(facebook_url, 'Facebook')
1937
1938         # Look for embedded VK player
1939         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1940         if mobj is not None:
1941             return self.url_result(mobj.group('url'), 'VK')
1942
1943         # Look for embedded Odnoklassniki player
1944         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1', webpage)
1945         if mobj is not None:
1946             return self.url_result(mobj.group('url'), 'Odnoklassniki')
1947
1948         # Look for embedded ivi player
1949         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1950         if mobj is not None:
1951             return self.url_result(mobj.group('url'), 'Ivi')
1952
1953         # Look for embedded Huffington Post player
1954         mobj = re.search(
1955             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1956         if mobj is not None:
1957             return self.url_result(mobj.group('url'), 'HuffPost')
1958
1959         # Look for embed.ly
1960         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1961         if mobj is not None:
1962             return self.url_result(mobj.group('url'))
1963         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1964         if mobj is not None:
1965             return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
1966
1967         # Look for funnyordie embed
1968         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1969         if matches:
1970             return _playlist_from_matches(
1971                 matches, getter=unescapeHTML, ie='FunnyOrDie')
1972
1973         # Look for BBC iPlayer embed
1974         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1975         if matches:
1976             return _playlist_from_matches(matches, ie='BBCCoUk')
1977
1978         # Look for embedded RUTV player
1979         rutv_url = RUTVIE._extract_url(webpage)
1980         if rutv_url:
1981             return self.url_result(rutv_url, 'RUTV')
1982
1983         # Look for embedded TVC player
1984         tvc_url = TVCIE._extract_url(webpage)
1985         if tvc_url:
1986             return self.url_result(tvc_url, 'TVC')
1987
1988         # Look for embedded SportBox player
1989         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1990         if sportbox_urls:
1991             return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1992
1993         # Look for embedded PornHub player
1994         pornhub_url = PornHubIE._extract_url(webpage)
1995         if pornhub_url:
1996             return self.url_result(pornhub_url, 'PornHub')
1997
1998         # Look for embedded XHamster player
1999         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
2000         if xhamster_urls:
2001             return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
2002
2003         # Look for embedded TNAFlixNetwork player
2004         tnaflix_urls = TNAFlixNetworkEmbedIE._extract_urls(webpage)
2005         if tnaflix_urls:
2006             return _playlist_from_matches(tnaflix_urls, ie=TNAFlixNetworkEmbedIE.ie_key())
2007
2008         # Look for embedded Tvigle player
2009         mobj = re.search(
2010             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
2011         if mobj is not None:
2012             return self.url_result(mobj.group('url'), 'Tvigle')
2013
2014         # Look for embedded TED player
2015         mobj = re.search(
2016             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
2017         if mobj is not None:
2018             return self.url_result(mobj.group('url'), 'TED')
2019
2020         # Look for embedded Ustream videos
2021         mobj = re.search(
2022             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
2023         if mobj is not None:
2024             return self.url_result(mobj.group('url'), 'Ustream')
2025
2026         # Look for embedded arte.tv player
2027         mobj = re.search(
2028             r'<(?:script|iframe) [^>]*?src="(?P<url>http://www\.arte\.tv/(?:playerv2/embed|arte_vp/index)[^"]+)"',
2029             webpage)
2030         if mobj is not None:
2031             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
2032
2033         # Look for embedded francetv player
2034         mobj = re.search(
2035             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
2036             webpage)
2037         if mobj is not None:
2038             return self.url_result(mobj.group('url'))
2039
2040         # Look for embedded smotri.com player
2041         smotri_url = SmotriIE._extract_url(webpage)
2042         if smotri_url:
2043             return self.url_result(smotri_url, 'Smotri')
2044
2045         # Look for embedded Myvi.ru player
2046         myvi_url = MyviIE._extract_url(webpage)
2047         if myvi_url:
2048             return self.url_result(myvi_url)
2049
2050         # Look for embedded soundcloud player
2051         soundcloud_urls = SoundcloudIE._extract_urls(webpage)
2052         if soundcloud_urls:
2053             return _playlist_from_matches(soundcloud_urls, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
2054
2055         # Look for embedded mtvservices player
2056         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
2057         if mtvservices_url:
2058             return self.url_result(mtvservices_url, ie='MTVServicesEmbedded')
2059
2060         # Look for embedded yahoo player
2061         mobj = re.search(
2062             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
2063             webpage)
2064         if mobj is not None:
2065             return self.url_result(mobj.group('url'), 'Yahoo')
2066
2067         # Look for embedded sbs.com.au player
2068         mobj = re.search(
2069             r'''(?x)
2070             (?:
2071                 <meta\s+property="og:video"\s+content=|
2072                 <iframe[^>]+?src=
2073             )
2074             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
2075             webpage)
2076         if mobj is not None:
2077             return self.url_result(mobj.group('url'), 'SBS')
2078
2079         # Look for embedded Cinchcast player
2080         mobj = re.search(
2081             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
2082             webpage)
2083         if mobj is not None:
2084             return self.url_result(mobj.group('url'), 'Cinchcast')
2085
2086         mobj = re.search(
2087             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
2088             webpage)
2089         if not mobj:
2090             mobj = re.search(
2091                 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
2092                 webpage)
2093         if mobj is not None:
2094             return self.url_result(mobj.group('url'), 'MLB')
2095
2096         mobj = re.search(
2097             r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
2098             webpage)
2099         if mobj is not None:
2100             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
2101
2102         mobj = re.search(
2103             r'<iframe[^>]+src="(?P<url>https?://(?:new\.)?livestream\.com/[^"]+/player[^"]+)"',
2104             webpage)
2105         if mobj is not None:
2106             return self.url_result(mobj.group('url'), 'Livestream')
2107
2108         # Look for Zapiks embed
2109         mobj = re.search(
2110             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
2111         if mobj is not None:
2112             return self.url_result(mobj.group('url'), 'Zapiks')
2113
2114         # Look for Kaltura embeds
2115         kaltura_url = KalturaIE._extract_url(webpage)
2116         if kaltura_url:
2117             return self.url_result(smuggle_url(kaltura_url, {'source_url': url}), KalturaIE.ie_key())
2118
2119         # Look for Eagle.Platform embeds
2120         eagleplatform_url = EaglePlatformIE._extract_url(webpage)
2121         if eagleplatform_url:
2122             return self.url_result(eagleplatform_url, EaglePlatformIE.ie_key())
2123
2124         # Look for ClipYou (uses Eagle.Platform) embeds
2125         mobj = re.search(
2126             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
2127         if mobj is not None:
2128             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
2129
2130         # Look for Pladform embeds
2131         pladform_url = PladformIE._extract_url(webpage)
2132         if pladform_url:
2133             return self.url_result(pladform_url)
2134
2135         # Look for Videomore embeds
2136         videomore_url = VideomoreIE._extract_url(webpage)
2137         if videomore_url:
2138             return self.url_result(videomore_url)
2139
2140         # Look for Playwire embeds
2141         mobj = re.search(
2142             r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
2143         if mobj is not None:
2144             return self.url_result(mobj.group('url'))
2145
2146         # Look for 5min embeds
2147         mobj = re.search(
2148             r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
2149         if mobj is not None:
2150             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
2151
2152         # Look for Crooks and Liars embeds
2153         mobj = re.search(
2154             r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
2155         if mobj is not None:
2156             return self.url_result(mobj.group('url'))
2157
2158         # Look for NBC Sports VPlayer embeds
2159         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
2160         if nbc_sports_url:
2161             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
2162
2163         # Look for NBC News embeds
2164         nbc_news_embed_url = re.search(
2165             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//www\.nbcnews\.com/widget/video-embed/[^"\']+)\1', webpage)
2166         if nbc_news_embed_url:
2167             return self.url_result(nbc_news_embed_url.group('url'), 'NBCNews')
2168
2169         # Look for Google Drive embeds
2170         google_drive_url = GoogleDriveIE._extract_url(webpage)
2171         if google_drive_url:
2172             return self.url_result(google_drive_url, 'GoogleDrive')
2173
2174         # Look for UDN embeds
2175         mobj = re.search(
2176             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage)
2177         if mobj is not None:
2178             return self.url_result(
2179                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
2180
2181         # Look for Senate ISVP iframe
2182         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
2183         if senate_isvp_url:
2184             return self.url_result(senate_isvp_url, 'SenateISVP')
2185
2186         # Look for Dailymotion Cloud videos
2187         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
2188         if dmcloud_url:
2189             return self.url_result(dmcloud_url, 'DailymotionCloud')
2190
2191         # Look for OnionStudios embeds
2192         onionstudios_url = OnionStudiosIE._extract_url(webpage)
2193         if onionstudios_url:
2194             return self.url_result(onionstudios_url)
2195
2196         # Look for ViewLift embeds
2197         viewlift_url = ViewLiftEmbedIE._extract_url(webpage)
2198         if viewlift_url:
2199             return self.url_result(viewlift_url)
2200
2201         # Look for JWPlatform embeds
2202         jwplatform_url = JWPlatformIE._extract_url(webpage)
2203         if jwplatform_url:
2204             return self.url_result(jwplatform_url, 'JWPlatform')
2205
2206         # Look for ScreenwaveMedia embeds
2207         mobj = re.search(ScreenwaveMediaIE.EMBED_PATTERN, webpage)
2208         if mobj is not None:
2209             return self.url_result(unescapeHTML(mobj.group('url')), 'ScreenwaveMedia')
2210
2211         # Look for Digiteka embeds
2212         digiteka_url = DigitekaIE._extract_url(webpage)
2213         if digiteka_url:
2214             return self.url_result(self._proto_relative_url(digiteka_url), DigitekaIE.ie_key())
2215
2216         # Look for Arkena embeds
2217         arkena_url = ArkenaIE._extract_url(webpage)
2218         if arkena_url:
2219             return self.url_result(arkena_url, ArkenaIE.ie_key())
2220
2221         # Look for Limelight embeds
2222         mobj = re.search(r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})', webpage)
2223         if mobj:
2224             lm = {
2225                 'Media': 'media',
2226                 'Channel': 'channel',
2227                 'ChannelList': 'channel_list',
2228             }
2229             return self.url_result('limelight:%s:%s' % (
2230                 lm[mobj.group(1)], mobj.group(2)), 'Limelight%s' % mobj.group(1), mobj.group(2))
2231
2232         # Look for AdobeTVVideo embeds
2233         mobj = re.search(
2234             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
2235             webpage)
2236         if mobj is not None:
2237             return self.url_result(
2238                 self._proto_relative_url(unescapeHTML(mobj.group(1))),
2239                 'AdobeTVVideo')
2240
2241         # Look for Vine embeds
2242         mobj = re.search(
2243             r'<iframe[^>]+src=[\'"]((?:https?:)?//(?:www\.)?vine\.co/v/[^/]+/embed/(?:simple|postcard))',
2244             webpage)
2245         if mobj is not None:
2246             return self.url_result(
2247                 self._proto_relative_url(unescapeHTML(mobj.group(1))), 'Vine')
2248
2249         # Look for VODPlatform embeds
2250         mobj = re.search(
2251             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vod-platform\.net/[eE]mbed/.+?)\1',
2252             webpage)
2253         if mobj is not None:
2254             return self.url_result(
2255                 self._proto_relative_url(unescapeHTML(mobj.group('url'))), 'VODPlatform')
2256
2257         # Look for Mangomolo embeds
2258         mobj = re.search(
2259             r'''(?x)<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?admin\.mangomolo.com/analytics/index\.php/customers/embed/
2260                 (?:
2261                     video\?.*?\bid=(?P<video_id>\d+)|
2262                     index\?.*?\bchannelid=(?P<channel_id>(?:[A-Za-z0-9+/=]|%2B|%2F|%3D)+)
2263                 ).+?)\1''', webpage)
2264         if mobj is not None:
2265             info = {
2266                 '_type': 'url_transparent',
2267                 'url': self._proto_relative_url(unescapeHTML(mobj.group('url'))),
2268                 'title': video_title,
2269                 'description': video_description,
2270                 'thumbnail': video_thumbnail,
2271                 'uploader': video_uploader,
2272             }
2273             video_id = mobj.group('video_id')
2274             if video_id:
2275                 info.update({
2276                     'ie_key': 'MangomoloVideo',
2277                     'id': video_id,
2278                 })
2279             else:
2280                 info.update({
2281                     'ie_key': 'MangomoloLive',
2282                     'id': mobj.group('channel_id'),
2283                 })
2284             return info
2285
2286         # Look for Instagram embeds
2287         instagram_embed_url = InstagramIE._extract_embed_url(webpage)
2288         if instagram_embed_url is not None:
2289             return self.url_result(
2290                 self._proto_relative_url(instagram_embed_url), InstagramIE.ie_key())
2291
2292         # Look for LiveLeak embeds
2293         liveleak_url = LiveLeakIE._extract_url(webpage)
2294         if liveleak_url:
2295             return self.url_result(liveleak_url, 'LiveLeak')
2296
2297         # Look for 3Q SDN embeds
2298         threeqsdn_url = ThreeQSDNIE._extract_url(webpage)
2299         if threeqsdn_url:
2300             return {
2301                 '_type': 'url_transparent',
2302                 'ie_key': ThreeQSDNIE.ie_key(),
2303                 'url': self._proto_relative_url(threeqsdn_url),
2304                 'title': video_title,
2305                 'description': video_description,
2306                 'thumbnail': video_thumbnail,
2307                 'uploader': video_uploader,
2308             }
2309
2310         # Look for VBOX7 embeds
2311         vbox7_url = Vbox7IE._extract_url(webpage)
2312         if vbox7_url:
2313             return self.url_result(vbox7_url, Vbox7IE.ie_key())
2314
2315         # Look for DBTV embeds
2316         dbtv_urls = DBTVIE._extract_urls(webpage)
2317         if dbtv_urls:
2318             return _playlist_from_matches(dbtv_urls, ie=DBTVIE.ie_key())
2319
2320         # Looking for http://schema.org/VideoObject
2321         json_ld = self._search_json_ld(
2322             webpage, video_id, default={}, expected_type='VideoObject')
2323         if json_ld.get('url'):
2324             info_dict.update({
2325                 'title': video_title or info_dict['title'],
2326                 'description': video_description,
2327                 'thumbnail': video_thumbnail,
2328                 'age_limit': age_limit
2329             })
2330             info_dict.update(json_ld)
2331             return info_dict
2332
2333         def check_video(vurl):
2334             if YoutubeIE.suitable(vurl):
2335                 return True
2336             vpath = compat_urlparse.urlparse(vurl).path
2337             vext = determine_ext(vpath)
2338             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
2339
2340         def filter_video(urls):
2341             return list(filter(check_video, urls))
2342
2343         # Start with something easy: JW Player in SWFObject
2344         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
2345         if not found:
2346             # Look for gorilla-vid style embedding
2347             found = filter_video(re.findall(r'''(?sx)
2348                 (?:
2349                     jw_plugins|
2350                     JWPlayerOptions|
2351                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
2352                 )
2353                 .*?
2354                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
2355         if not found:
2356             # Broaden the search a little bit
2357             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
2358         if not found:
2359             # Broaden the findall a little bit: JWPlayer JS loader
2360             found = filter_video(re.findall(
2361                 r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
2362         if not found:
2363             # Flow player
2364             found = filter_video(re.findall(r'''(?xs)
2365                 flowplayer\("[^"]+",\s*
2366                     \{[^}]+?\}\s*,
2367                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
2368                         ["']?url["']?\s*:\s*["']([^"']+)["']
2369             ''', webpage))
2370         if not found:
2371             # Cinerama player
2372             found = re.findall(
2373                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
2374         if not found:
2375             # Try to find twitter cards info
2376             # twitter:player:stream should be checked before twitter:player since
2377             # it is expected to contain a raw stream (see
2378             # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser)
2379             found = filter_video(re.findall(
2380                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
2381         if not found:
2382             # We look for Open Graph info:
2383             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
2384             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
2385             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
2386             if m_video_type is not None:
2387                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
2388         if not found:
2389             # HTML5 video
2390             found = re.findall(r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
2391         if not found:
2392             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
2393             found = re.search(
2394                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
2395                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
2396                 webpage)
2397             if not found:
2398                 # Look also in Refresh HTTP header
2399                 refresh_header = head_response.headers.get('Refresh')
2400                 if refresh_header:
2401                     # In python 2 response HTTP headers are bytestrings
2402                     if sys.version_info < (3, 0) and isinstance(refresh_header, str):
2403                         refresh_header = refresh_header.decode('iso-8859-1')
2404                     found = re.search(REDIRECT_REGEX, refresh_header)
2405             if found:
2406                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
2407                 self.report_following_redirect(new_url)
2408                 return {
2409                     '_type': 'url',
2410                     'url': new_url,
2411                 }
2412
2413         if not found:
2414             # twitter:player is a https URL to iframe player that may or may not
2415             # be supported by youtube-dl thus this is checked the very last (see
2416             # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser)
2417             embed_url = self._html_search_meta('twitter:player', webpage, default=None)
2418             if embed_url:
2419                 return self.url_result(embed_url)
2420
2421         if not found:
2422             raise UnsupportedError(url)
2423
2424         entries = []
2425         for video_url in orderedSet(found):
2426             video_url = unescapeHTML(video_url)
2427             video_url = video_url.replace('\\/', '/')
2428             video_url = compat_urlparse.urljoin(url, video_url)
2429             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
2430
2431             # Sometimes, jwplayer extraction will result in a YouTube URL
2432             if YoutubeIE.suitable(video_url):
2433                 entries.append(self.url_result(video_url, 'Youtube'))
2434                 continue
2435
2436             # here's a fun little line of code for you:
2437             video_id = os.path.splitext(video_id)[0]
2438
2439             entry_info_dict = {
2440                 'id': video_id,
2441                 'uploader': video_uploader,
2442                 'title': video_title,
2443                 'age_limit': age_limit,
2444             }
2445
2446             ext = determine_ext(video_url)
2447             if ext == 'smil':
2448                 entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
2449             elif ext == 'xspf':
2450                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
2451             elif ext == 'm3u8':
2452                 entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
2453             elif ext == 'mpd':
2454                 entry_info_dict['formats'] = self._extract_mpd_formats(video_url, video_id)
2455             elif ext == 'f4m':
2456                 entry_info_dict['formats'] = self._extract_f4m_formats(video_url, video_id)
2457             else:
2458                 entry_info_dict['url'] = video_url
2459
2460             if entry_info_dict.get('formats'):
2461                 self._sort_formats(entry_info_dict['formats'])
2462
2463             entries.append(entry_info_dict)
2464
2465         if len(entries) == 1:
2466             return entries[0]
2467         else:
2468             for num, e in enumerate(entries, start=1):
2469                 # 'url' results don't have a title
2470                 if e.get('title') is not None:
2471                     e['title'] = '%s (%d)' % (e['title'], num)
2472             return {
2473                 '_type': 'playlist',
2474                 'entries': entries,
2475             }