_ 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 .vimeo import VimeoIE
  51 from .dailymotion import DailymotionCloudIE
  52 from .onionstudios import OnionStudiosIE
  53 from .snagfilms import SnagFilmsEmbedIE
  54 from .screenwavemedia import ScreenwaveMediaIE
  55 from .mtv import MTVServicesEmbeddedIE
  56
  57
  58 class GenericIE(InfoExtractor):
  59     IE_DESC = 'Generic downloader that works on some sites'
  60     _VALID_URL = r'.*'
  61     IE_NAME = 'generic'
  62     _TESTS = [
  63         # Direct link to a video
  64         {
  65             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  66             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  67             'info_dict': {
  68                 'id': 'trailer',
  69                 'ext': 'mp4',
  70                 'title': 'trailer',
  71                 'upload_date': '20100513',
  72             }
  73         },
  74         # Direct link to media delivered compressed (until Accept-Encoding is *)
  75         {
  76             'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
  77             'md5': '128c42e68b13950268b648275386fc74',
  78             'info_dict': {
  79                 'id': 'FictionJunction-Parallel_Hearts',
  80                 'ext': 'flac',
  81                 'title': 'FictionJunction-Parallel_Hearts',
  82                 'upload_date': '20140522',
  83             },
  84             'expected_warnings': [
  85                 'URL could be a direct video link, returning it as such.'
  86             ]
  87         },
  88         # Direct download with broken HEAD
  89         {
  90             'url': 'http://ai-radio.org:8000/radio.opus',
  91             'info_dict': {
  92                 'id': 'radio',
  93                 'ext': 'opus',
  94                 'title': 'radio',
  95             },
  96             'params': {
  97                 'skip_download': True,  # infinite live stream
  98             },
  99             'expected_warnings': [
 100                 r'501.*Not Implemented'
 101             ],
 102         },
 103         # Direct link with incorrect MIME type
 104         {
 105             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 106             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 107             'info_dict': {
 108                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 109                 'id': '5_Lennart_Poettering_-_Systemd',
 110                 'ext': 'webm',
 111                 'title': '5_Lennart_Poettering_-_Systemd',
 112                 'upload_date': '20141120',
 113             },
 114             'expected_warnings': [
 115                 'URL could be a direct video link, returning it as such.'
 116             ]
 117         },
 118         # RSS feed
 119         {
 120             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 121             'info_dict': {
 122                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 123                 'title': 'Zero Punctuation',
 124                 'description': 're:.*groundbreaking video review series.*'
 125             },
 126             'playlist_mincount': 11,
 127         },
 128         # RSS feed with enclosure
 129         {
 130             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 131             'info_dict': {
 132                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 133                 'ext': 'm4v',
 134                 'upload_date': '20150228',
 135                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 136             }
 137         },
 138         # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
 139         {
 140             'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
 141             'info_dict': {
 142                 'id': 'smil',
 143                 'ext': 'mp4',
 144                 'title': 'Automatics, robotics and biocybernetics',
 145                 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
 146                 'upload_date': '20130627',
 147                 'formats': 'mincount:16',
 148                 'subtitles': 'mincount:1',
 149             },
 150             'params': {
 151                 'force_generic_extractor': True,
 152                 'skip_download': True,
 153             },
 154         },
 155         # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html
 156         {
 157             'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil',
 158             'info_dict': {
 159                 'id': 'hds',
 160                 'ext': 'flv',
 161                 'title': 'hds',
 162                 'formats': 'mincount:1',
 163             },
 164             'params': {
 165                 'skip_download': True,
 166             },
 167         },
 168         # SMIL from https://www.restudy.dk/video/play/id/1637
 169         {
 170             'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml',
 171             'info_dict': {
 172                 'id': 'video_1637',
 173                 'ext': 'flv',
 174                 'title': 'video_1637',
 175                 'formats': 'mincount:3',
 176             },
 177             'params': {
 178                 'skip_download': True,
 179             },
 180         },
 181         # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm
 182         {
 183             'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil',
 184             'info_dict': {
 185                 'id': 'smil-service',
 186                 'ext': 'flv',
 187                 'title': 'smil-service',
 188                 'formats': 'mincount:1',
 189             },
 190             'params': {
 191                 'skip_download': True,
 192             },
 193         },
 194         # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370
 195         {
 196             'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil',
 197             'info_dict': {
 198                 'id': '4719370',
 199                 'ext': 'mp4',
 200                 'title': '571de1fd-47bc-48db-abf9-238872a58d1f',
 201                 'formats': 'mincount:3',
 202             },
 203             'params': {
 204                 'skip_download': True,
 205             },
 206         },
 207         # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html
 208         {
 209             'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf',
 210             'info_dict': {
 211                 'id': 'mZlp2ctYIUEB',
 212                 'ext': 'mp4',
 213                 'title': 'Tikibad ontruimd wegens brand',
 214                 'description': 'md5:05ca046ff47b931f9b04855015e163a4',
 215                 'thumbnail': 're:^https?://.*\.jpg$',
 216                 'duration': 33,
 217             },
 218             'params': {
 219                 'skip_download': True,
 220             },
 221         },
 222         # google redirect
 223         {
 224             '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',
 225             'info_dict': {
 226                 'id': 'cmQHVoWB5FY',
 227                 'ext': 'mp4',
 228                 'upload_date': '20130224',
 229                 'uploader_id': 'TheVerge',
 230                 'description': 're:^Chris Ziegler takes a look at the\.*',
 231                 'uploader': 'The Verge',
 232                 'title': 'First Firefox OS phones side-by-side',
 233             },
 234             'params': {
 235                 'skip_download': False,
 236             }
 237         },
 238         {
 239             # redirect in Refresh HTTP header
 240             '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',
 241             'info_dict': {
 242                 'id': 'pO8h3EaFRdo',
 243                 'ext': 'mp4',
 244                 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
 245                 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5',
 246                 'upload_date': '20150917',
 247                 'uploader_id': 'brtvofficial',
 248                 'uploader': 'Boiler Room',
 249             },
 250             'params': {
 251                 'skip_download': False,
 252             },
 253         },
 254         {
 255             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
 256             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
 257             'info_dict': {
 258                 'id': '13601338388002',
 259                 'ext': 'mp4',
 260                 'uploader': 'www.hodiho.fr',
 261                 'title': 'R\u00e9gis plante sa Jeep',
 262             }
 263         },
 264         # bandcamp page with custom domain
 265         {
 266             'add_ie': ['Bandcamp'],
 267             'url': 'http://bronyrock.com/track/the-pony-mash',
 268             'info_dict': {
 269                 'id': '3235767654',
 270                 'ext': 'mp3',
 271                 'title': 'The Pony Mash',
 272                 'uploader': 'M_Pallante',
 273             },
 274             'skip': 'There is a limit of 200 free downloads / month for the test song',
 275         },
 276         # embedded brightcove video
 277         # it also tests brightcove videos that need to set the 'Referer' in the
 278         # http requests
 279         {
 280             'add_ie': ['BrightcoveLegacy'],
 281             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
 282             'info_dict': {
 283                 'id': '2765128793001',
 284                 'ext': 'mp4',
 285                 'title': 'Le cours de bourse : l’analyse technique',
 286                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
 287                 'uploader': 'BFM BUSINESS',
 288             },
 289             'params': {
 290                 'skip_download': True,
 291             },
 292         },
 293         {
 294             # https://github.com/rg3/youtube-dl/issues/2253
 295             'url': 'http://bcove.me/i6nfkrc3',
 296             'md5': '0ba9446db037002366bab3b3eb30c88c',
 297             'info_dict': {
 298                 'id': '3101154703001',
 299                 'ext': 'mp4',
 300                 'title': 'Still no power',
 301                 'uploader': 'thestar.com',
 302                 '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.',
 303             },
 304             'add_ie': ['BrightcoveLegacy'],
 305         },
 306         {
 307             'url': 'http://www.championat.com/video/football/v/87/87499.html',
 308             'md5': 'fb973ecf6e4a78a67453647444222983',
 309             'info_dict': {
 310                 'id': '3414141473001',
 311                 'ext': 'mp4',
 312                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 313                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 314                 'uploader': 'Championat',
 315             },
 316         },
 317         {
 318             # https://github.com/rg3/youtube-dl/issues/3541
 319             'add_ie': ['BrightcoveLegacy'],
 320             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 321             'info_dict': {
 322                 'id': '3866516442001',
 323                 'ext': 'mp4',
 324                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 325                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 326                 'uploader': 'SBS Broadcasting',
 327             },
 328             'skip': 'Restricted to Netherlands',
 329             'params': {
 330                 'skip_download': True,  # m3u8 download
 331             },
 332         },
 333         # ooyala video
 334         {
 335             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 336             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 337             'info_dict': {
 338                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 339                 'ext': 'mp4',
 340                 'title': '2cc213299525360.mov',  # that's what we get
 341             },
 342             'add_ie': ['Ooyala'],
 343         },
 344         {
 345             # ooyala video embedded with http://player.ooyala.com/iframe.js
 346             'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/',
 347             'info_dict': {
 348                 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB',
 349                 'ext': 'mp4',
 350                 'title': '"Steve Jobs: Man in the Machine" trailer',
 351                 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
 352             },
 353             'params': {
 354                 'skip_download': True,
 355             },
 356         },
 357         # multiple ooyala embeds on SBN network websites
 358         {
 359             'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 360             'info_dict': {
 361                 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 362                 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
 363             },
 364             'playlist_mincount': 3,
 365             'params': {
 366                 'skip_download': True,
 367             },
 368             'add_ie': ['Ooyala'],
 369         },
 370         # embed.ly video
 371         {
 372             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 373             'info_dict': {
 374                 'id': '9ODmcdjQcHQ',
 375                 'ext': 'mp4',
 376                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 377                 'upload_date': '20140225',
 378                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 379                 'uploader': 'Tested',
 380                 'uploader_id': 'testedcom',
 381             },
 382             # No need to test YoutubeIE here
 383             'params': {
 384                 'skip_download': True,
 385             },
 386         },
 387         # funnyordie embed
 388         {
 389             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 390             'info_dict': {
 391                 'id': '18e820ec3f',
 392                 'ext': 'mp4',
 393                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 394                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 395             },
 396         },
 397         # RUTV embed
 398         {
 399             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 400             'info_dict': {
 401                 'id': '776940',
 402                 'ext': 'mp4',
 403                 'title': 'Охотское море стало целиком российским',
 404                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 405             },
 406             'params': {
 407                 # m3u8 download
 408                 'skip_download': True,
 409             },
 410         },
 411         # TVC embed
 412         {
 413             '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/',
 414             'info_dict': {
 415                 'id': '55304',
 416                 'ext': 'mp4',
 417                 'title': 'Дошкольное воспитание',
 418             },
 419         },
 420         # SportBox embed
 421         {
 422             'url': 'http://www.vestifinance.ru/articles/25753',
 423             'info_dict': {
 424                 'id': '25753',
 425                 'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
 426             },
 427             'playlist': [{
 428                 'info_dict': {
 429                     'id': '370908',
 430                     'title': 'Госзаказ. День 3',
 431                     'ext': 'mp4',
 432                 }
 433             }, {
 434                 'info_dict': {
 435                     'id': '370905',
 436                     'title': 'Госзаказ. День 2',
 437                     'ext': 'mp4',
 438                 }
 439             }, {
 440                 'info_dict': {
 441                     'id': '370902',
 442                     'title': 'Госзаказ. День 1',
 443                     'ext': 'mp4',
 444                 }
 445             }],
 446             'params': {
 447                 # m3u8 download
 448                 'skip_download': True,
 449             },
 450         },
 451         # Myvi.ru embed
 452         {
 453             'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
 454             'info_dict': {
 455                 'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
 456                 'ext': 'mp4',
 457                 'title': 'Ужастики, русский трейлер (2015)',
 458                 'thumbnail': 're:^https?://.*\.jpg$',
 459                 'duration': 153,
 460             }
 461         },
 462         # XHamster embed
 463         {
 464             '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',
 465             'info_dict': {
 466                 'id': 'showthread',
 467                 'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
 468             },
 469             'playlist_mincount': 7,
 470         },
 471         # Embedded TED video
 472         {
 473             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 474             'md5': '65fdff94098e4a607385a60c5177c638',
 475             'info_dict': {
 476                 'id': '1969',
 477                 'ext': 'mp4',
 478                 'title': 'Hidden miracles of the natural world',
 479                 'uploader': 'Louie Schwartzberg',
 480                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 481             }
 482         },
 483         # Embeded Ustream video
 484         {
 485             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 486             'md5': '27b99cdb639c9b12a79bca876a073417',
 487             'info_dict': {
 488                 'id': '45734260',
 489                 'ext': 'flv',
 490                 'uploader': 'AU SPA:  The NSA and Privacy',
 491                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 492             }
 493         },
 494         # nowvideo embed hidden behind percent encoding
 495         {
 496             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 497             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 498             'info_dict': {
 499                 'id': '06e53103ca9aa',
 500                 'ext': 'flv',
 501                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 502                 'description': 'No description',
 503             },
 504         },
 505         # arte embed
 506         {
 507             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 508             'md5': '7653032cbb25bf6c80d80f217055fa43',
 509             'info_dict': {
 510                 'id': '048195-004_PLUS7-F',
 511                 'ext': 'flv',
 512                 'title': 'X:enius',
 513                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 514                 'upload_date': '20140320',
 515             },
 516             'params': {
 517                 'skip_download': 'Requires rtmpdump'
 518             }
 519         },
 520         # francetv embed
 521         {
 522             'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
 523             'info_dict': {
 524                 'id': 'EV_30231',
 525                 'ext': 'mp4',
 526                 'title': 'Alcaline, le concert avec Calogero',
 527                 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
 528                 'upload_date': '20150226',
 529                 'timestamp': 1424989860,
 530                 'duration': 5400,
 531             },
 532             'params': {
 533                 # m3u8 downloads
 534                 'skip_download': True,
 535             },
 536             'expected_warnings': [
 537                 'Forbidden'
 538             ]
 539         },
 540         # Condé Nast embed
 541         {
 542             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 543             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 544             'info_dict': {
 545                 'id': '53501be369702d3275860000',
 546                 'ext': 'mp4',
 547                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 548             }
 549         },
 550         # Dailymotion embed
 551         {
 552             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 553             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 554             'info_dict': {
 555                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 556                 'ext': 'mp4',
 557                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 558                 'uploader': 'Spi0n',
 559             },
 560             'add_ie': ['Dailymotion'],
 561         },
 562         # YouTube embed
 563         {
 564             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 565             'info_dict': {
 566                 'id': 'FXRb4ykk4S0',
 567                 'ext': 'mp4',
 568                 'title': 'The NBL Auction 2014',
 569                 'uploader': 'BADMINTON England',
 570                 'uploader_id': 'BADMINTONEvents',
 571                 'upload_date': '20140603',
 572                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 573             },
 574             'add_ie': ['Youtube'],
 575             'params': {
 576                 'skip_download': True,
 577             }
 578         },
 579         # MTVSercices embed
 580         {
 581             'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
 582             'md5': '35727f82f58c76d996fc188f9755b0d5',
 583             'info_dict': {
 584                 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
 585                 'ext': 'mp4',
 586                 'title': 'Review',
 587                 'description': 'Mario\'s life in the fast lane has never looked so good.',
 588             },
 589         },
 590         # YouTube embed via <data-embed-url="">
 591         {
 592             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 593             'info_dict': {
 594                 'id': '4vAffPZIT44',
 595                 'ext': 'mp4',
 596                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 597                 'uploader': 'Gameloft',
 598                 'uploader_id': 'gameloft',
 599                 'upload_date': '20140828',
 600                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 601             },
 602             'params': {
 603                 'skip_download': True,
 604             }
 605         },
 606         # Camtasia studio
 607         {
 608             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 609             'playlist': [{
 610                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 611                 'info_dict': {
 612                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 613                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 614                     'ext': 'flv',
 615                     'duration': 2235.90,
 616                 }
 617             }, {
 618                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 619                 'info_dict': {
 620                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 621                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 622                     'ext': 'flv',
 623                     'duration': 2235.93,
 624                 }
 625             }],
 626             'info_dict': {
 627                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 628             }
 629         },
 630         # Flowplayer
 631         {
 632             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 633             'md5': '9d65602bf31c6e20014319c7d07fba27',
 634             'info_dict': {
 635                 'id': '5123ea6d5e5a7',
 636                 'ext': 'mp4',
 637                 'age_limit': 18,
 638                 'uploader': 'www.handjobhub.com',
 639                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 640             }
 641         },
 642         # Multiple brightcove videos
 643         # https://github.com/rg3/youtube-dl/issues/2283
 644         {
 645             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 646             'info_dict': {
 647                 'id': 'always-never',
 648                 'title': 'Always / Never - The New Yorker',
 649             },
 650             'playlist_count': 3,
 651             'params': {
 652                 'extract_flat': False,
 653                 'skip_download': True,
 654             }
 655         },
 656         # MLB embed
 657         {
 658             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 659             'md5': '96f09a37e44da40dd083e12d9a683327',
 660             'info_dict': {
 661                 'id': '33322633',
 662                 'ext': 'mp4',
 663                 'title': 'Ump changes call to ball',
 664                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 665                 'duration': 48,
 666                 'timestamp': 1401537900,
 667                 'upload_date': '20140531',
 668                 'thumbnail': 're:^https?://.*\.jpg$',
 669             },
 670         },
 671         # Wistia embed
 672         {
 673             'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 674             'md5': '8788b683c777a5cf25621eaf286d0c23',
 675             'info_dict': {
 676                 'id': '1cfaf6b7ea',
 677                 'ext': 'mov',
 678                 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
 679                 'duration': 643.0,
 680                 'filesize': 182808282,
 681                 'uploader': 'education-portal.com',
 682             },
 683         },
 684         {
 685             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 686             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 687             'info_dict': {
 688                 'id': 'uxjb0lwrcz',
 689                 'ext': 'mp4',
 690                 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
 691                 'duration': 1715.0,
 692                 'uploader': 'thoughtworks.wistia.com',
 693             },
 694         },
 695         # Soundcloud embed
 696         {
 697             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 698             'info_dict': {
 699                 'id': '174391317',
 700                 'ext': 'mp3',
 701                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 702                 'uploader': 'Sophos Security',
 703                 'title': 'Chet Chat 171 - Oct 29, 2014',
 704                 'upload_date': '20141029',
 705             }
 706         },
 707         # Livestream embed
 708         {
 709             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 710             'info_dict': {
 711                 'id': '67864563',
 712                 'ext': 'flv',
 713                 'upload_date': '20141112',
 714                 'title': 'Rosetta #CometLanding webcast HL 10',
 715             }
 716         },
 717         # LazyYT
 718         {
 719             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
 720             'info_dict': {
 721                 'id': '1986',
 722                 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
 723             },
 724             'playlist_mincount': 2,
 725         },
 726         # Cinchcast embed
 727         {
 728             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
 729             'info_dict': {
 730                 'id': '7141703',
 731                 'ext': 'mp3',
 732                 'upload_date': '20141126',
 733                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
 734             }
 735         },
 736         # Cinerama player
 737         {
 738             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
 739             'info_dict': {
 740                 'id': '730m_DandD_1901_512k',
 741                 'ext': 'mp4',
 742                 'uploader': 'www.abc.net.au',
 743                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
 744             }
 745         },
 746         # embedded viddler video
 747         {
 748             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
 749             'info_dict': {
 750                 'id': '4d03aad9',
 751                 'ext': 'mp4',
 752                 'uploader': 'deadspin',
 753                 'title': 'WALL-TO-GORTAT',
 754                 'timestamp': 1422285291,
 755                 'upload_date': '20150126',
 756             },
 757             'add_ie': ['Viddler'],
 758         },
 759         # Libsyn embed
 760         {
 761             'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
 762             'info_dict': {
 763                 'id': '3377616',
 764                 'ext': 'mp3',
 765                 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
 766                 'description': 'md5:601cb790edd05908957dae8aaa866465',
 767                 'upload_date': '20150220',
 768             },
 769         },
 770         # jwplayer YouTube
 771         {
 772             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
 773             'info_dict': {
 774                 'id': 'Mrj4DVp2zeA',
 775                 'ext': 'mp4',
 776                 'upload_date': '20150212',
 777                 'uploader': 'The National Archives UK',
 778                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
 779                 'uploader_id': 'NationalArchives08',
 780                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
 781             },
 782         },
 783         # rtl.nl embed
 784         {
 785             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
 786             'playlist_mincount': 5,
 787             'info_dict': {
 788                 'id': 'aanslagen-kopenhagen',
 789                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
 790             }
 791         },
 792         # Zapiks embed
 793         {
 794             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
 795             'info_dict': {
 796                 'id': '118046',
 797                 'ext': 'mp4',
 798                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
 799             }
 800         },
 801         # Kaltura embed
 802         {
 803             'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
 804             'info_dict': {
 805                 'id': '1_eergr3h1',
 806                 'ext': 'mp4',
 807                 'upload_date': '20150226',
 808                 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
 809                 'timestamp': int,
 810                 'title': 'John Carlson Postgame 2/25/15',
 811             },
 812         },
 813         # Kaltura embed (different embed code)
 814         {
 815             'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
 816             'info_dict': {
 817                 'id': '1_a52wc67y',
 818                 'ext': 'flv',
 819                 'upload_date': '20150127',
 820                 'uploader_id': 'PremierMedia',
 821                 'timestamp': int,
 822                 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
 823             },
 824         },
 825         # Kaltura embed protected with referrer
 826         {
 827             'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero',
 828             'info_dict': {
 829                 'id': '1_g4fbemnq',
 830                 'ext': 'mp4',
 831                 'title': 'Violetta - Achter De Schermen - Ruggero',
 832                 'description': 'Achter de schermen met Ruggero',
 833                 'timestamp': 1435133761,
 834                 'upload_date': '20150624',
 835                 'uploader_id': 'echojecka',
 836             },
 837         },
 838         # Eagle.Platform embed (generic URL)
 839         {
 840             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
 841             'info_dict': {
 842                 'id': '227304',
 843                 'ext': 'mp4',
 844                 'title': 'Навальный вышел на свободу',
 845                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
 846                 'thumbnail': 're:^https?://.*\.jpg$',
 847                 'duration': 87,
 848                 'view_count': int,
 849                 'age_limit': 0,
 850             },
 851         },
 852         # ClipYou (Eagle.Platform) embed (custom URL)
 853         {
 854             'url': 'http://muz-tv.ru/play/7129/',
 855             'info_dict': {
 856                 'id': '12820',
 857                 'ext': 'mp4',
 858                 'title': "'O Sole Mio",
 859                 'thumbnail': 're:^https?://.*\.jpg$',
 860                 'duration': 216,
 861                 'view_count': int,
 862             },
 863         },
 864         # Pladform embed
 865         {
 866             'url': 'http://muz-tv.ru/kinozal/view/7400/',
 867             'info_dict': {
 868                 'id': '100183293',
 869                 'ext': 'mp4',
 870                 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
 871                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
 872                 'thumbnail': 're:^https?://.*\.jpg$',
 873                 'duration': 694,
 874                 'age_limit': 0,
 875             },
 876         },
 877         # Playwire embed
 878         {
 879             'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
 880             'info_dict': {
 881                 'id': '3519514',
 882                 'ext': 'mp4',
 883                 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
 884                 'thumbnail': 're:^https?://.*\.png$',
 885                 'duration': 45.115,
 886             },
 887         },
 888         # 5min embed
 889         {
 890             'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
 891             'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
 892             'info_dict': {
 893                 'id': '518726732',
 894                 'ext': 'mp4',
 895                 'title': 'Facebook Creates "On This Day" | Crunch Report',
 896             },
 897         },
 898         # SVT embed
 899         {
 900             'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 901             'info_dict': {
 902                 'id': '2900353',
 903                 'ext': 'flv',
 904                 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
 905                 'duration': 27,
 906                 'age_limit': 0,
 907             },
 908         },
 909         # Crooks and Liars embed
 910         {
 911             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
 912             'info_dict': {
 913                 'id': '8RUoRhRi',
 914                 'ext': 'mp4',
 915                 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
 916                 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
 917                 'timestamp': 1428207000,
 918                 'upload_date': '20150405',
 919                 'uploader': 'Heather',
 920             },
 921         },
 922         # Crooks and Liars external embed
 923         {
 924             'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
 925             'info_dict': {
 926                 'id': 'MTE3MjUtMzQ2MzA',
 927                 'ext': 'mp4',
 928                 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
 929                 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
 930                 'timestamp': 1265032391,
 931                 'upload_date': '20100201',
 932                 'uploader': 'Heather',
 933             },
 934         },
 935         # NBC Sports vplayer embed
 936         {
 937             'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
 938             'info_dict': {
 939                 'id': 'ln7x1qSThw4k',
 940                 'ext': 'flv',
 941                 'title': "PFT Live: New leader in the 'new-look' defense",
 942                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
 943             },
 944         },
 945         # UDN embed
 946         {
 947             'url': 'http://www.udn.com/news/story/7314/822787',
 948             'md5': 'fd2060e988c326991037b9aff9df21a6',
 949             'info_dict': {
 950                 'id': '300346',
 951                 'ext': 'mp4',
 952                 'title': '中一中男師變性 全校師生力挺',
 953                 'thumbnail': 're:^https?://.*\.jpg$',
 954             }
 955         },
 956         # Ooyala embed
 957         {
 958             'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
 959             'info_dict': {
 960                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
 961                 'ext': 'mp4',
 962                 'description': 'VIDEO: Index/Match versus VLOOKUP.',
 963                 'title': 'This is what separates the Excel masters from the wannabes',
 964             },
 965             'params': {
 966                 # m3u8 downloads
 967                 'skip_download': True,
 968             }
 969         },
 970         # Contains a SMIL manifest
 971         {
 972             'url': 'http://www.telewebion.com/fa/1263668/%D9%82%D8%B1%D8%B9%D9%87%E2%80%8C%DA%A9%D8%B4%DB%8C-%D9%84%DB%8C%DA%AF-%D9%82%D9%87%D8%B1%D9%85%D8%A7%D9%86%D8%A7%D9%86-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7/%2B-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84.html',
 973             'info_dict': {
 974                 'id': 'file',
 975                 'ext': 'flv',
 976                 'title': '+ Football: Lottery Champions League Europe',
 977                 'uploader': 'www.telewebion.com',
 978             },
 979             'params': {
 980                 # rtmpe downloads
 981                 'skip_download': True,
 982             }
 983         },
 984         # Brightcove URL in single quotes
 985         {
 986             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
 987             'md5': '4ae374f1f8b91c889c4b9203c8c752af',
 988             'info_dict': {
 989                 'id': '4255764656001',
 990                 'ext': 'mp4',
 991                 'title': 'SN Presents: Russell Martin, World Citizen',
 992                 '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.',
 993                 'uploader': 'Rogers Sportsnet',
 994             },
 995         },
 996         # Dailymotion Cloud video
 997         {
 998             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
 999             'md5': '49444254273501a64675a7e68c502681',
1000             'info_dict': {
1001                 'id': '5585de919473990de4bee11b',
1002                 'ext': 'mp4',
1003                 'title': 'Le débat',
1004                 'thumbnail': 're:^https?://.*\.jpe?g$',
1005             }
1006         },
1007         # OnionStudios embed
1008         {
1009             'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
1010             'info_dict': {
1011                 'id': '2855',
1012                 'ext': 'mp4',
1013                 'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
1014                 'thumbnail': 're:^https?://.*\.jpe?g$',
1015                 'uploader': 'ClickHole',
1016                 'uploader_id': 'clickhole',
1017             }
1018         },
1019         # SnagFilms embed
1020         {
1021             'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
1022             'info_dict': {
1023                 'id': '74849a00-85a9-11e1-9660-123139220831',
1024                 'ext': 'mp4',
1025                 'title': '#whilewewatch',
1026             }
1027         },
1028         # AdobeTVVideo embed
1029         {
1030             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
1031             'md5': '43662b577c018ad707a63766462b1e87',
1032             'info_dict': {
1033                 'id': '2456',
1034                 'ext': 'mp4',
1035                 'title': 'New experience with Acrobat DC',
1036                 'description': 'New experience with Acrobat DC',
1037                 'duration': 248.667,
1038             },
1039         },
1040         # ScreenwaveMedia embed
1041         {
1042             'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1',
1043             'md5': '24ace5baba0d35d55c6810b51f34e9e0',
1044             'info_dict': {
1045                 'id': 'cinemasnob-55d26273809dd',
1046                 'ext': 'mp4',
1047                 'title': 'cinemasnob',
1048             },
1049         },
1050         # BrightcoveInPageEmbed embed
1051         {
1052             'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
1053             'info_dict': {
1054                 'id': '4238694884001',
1055                 'ext': 'flv',
1056                 'title': 'Tabletop: Dread, Last Thoughts',
1057                 'description': 'Tabletop: Dread, Last Thoughts',
1058                 'duration': 51690,
1059             },
1060         },
1061         # JWPlayer with M3U8
1062         {
1063             'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video',
1064             'info_dict': {
1065                 'id': 'playlist',
1066                 'ext': 'mp4',
1067                 'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ',
1068                 'uploader': 'ren.tv',
1069             },
1070             'params': {
1071                 # m3u8 downloads
1072                 'skip_download': True,
1073             }
1074         }
1075     ]
1076
1077     def report_following_redirect(self, new_url):
1078         """Report information extraction."""
1079         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
1080
1081     def _extract_rss(self, url, video_id, doc):
1082         playlist_title = doc.find('./channel/title').text
1083         playlist_desc_el = doc.find('./channel/description')
1084         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
1085
1086         entries = []
1087         for it in doc.findall('./channel/item'):
1088             next_url = xpath_text(it, 'link', fatal=False)
1089             if not next_url:
1090                 enclosure_nodes = it.findall('./enclosure')
1091                 for e in enclosure_nodes:
1092                     next_url = e.attrib.get('url')
1093                     if next_url:
1094                         break
1095
1096             if not next_url:
1097                 continue
1098
1099             entries.append({
1100                 '_type': 'url',
1101                 'url': next_url,
1102                 'title': it.find('title').text,
1103             })
1104
1105         return {
1106             '_type': 'playlist',
1107             'id': url,
1108             'title': playlist_title,
1109             'description': playlist_desc,
1110             'entries': entries,
1111         }
1112
1113     def _extract_camtasia(self, url, video_id, webpage):
1114         """ Returns None if no camtasia video can be found. """
1115
1116         camtasia_cfg = self._search_regex(
1117             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
1118             webpage, 'camtasia configuration file', default=None)
1119         if camtasia_cfg is None:
1120             return None
1121
1122         title = self._html_search_meta('DC.title', webpage, fatal=True)
1123
1124         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
1125         camtasia_cfg = self._download_xml(
1126             camtasia_url, video_id,
1127             note='Downloading camtasia configuration',
1128             errnote='Failed to download camtasia configuration')
1129         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
1130
1131         entries = []
1132         for n in fileset_node.getchildren():
1133             url_n = n.find('./uri')
1134             if url_n is None:
1135                 continue
1136
1137             entries.append({
1138                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
1139                 'title': '%s - %s' % (title, n.tag),
1140                 'url': compat_urlparse.urljoin(url, url_n.text),
1141                 'duration': float_or_none(n.find('./duration').text),
1142             })
1143
1144         return {
1145             '_type': 'playlist',
1146             'entries': entries,
1147             'title': title,
1148         }
1149
1150     def _real_extract(self, url):
1151         if url.startswith('//'):
1152             return {
1153                 '_type': 'url',
1154                 'url': self.http_scheme() + url,
1155             }
1156
1157         parsed_url = compat_urlparse.urlparse(url)
1158         if not parsed_url.scheme:
1159             default_search = self._downloader.params.get('default_search')
1160             if default_search is None:
1161                 default_search = 'fixup_error'
1162
1163             if default_search in ('auto', 'auto_warning', 'fixup_error'):
1164                 if '/' in url:
1165                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
1166                     return self.url_result('http://' + url)
1167                 elif default_search != 'fixup_error':
1168                     if default_search == 'auto_warning':
1169                         if re.match(r'^(?:url|URL)$', url):
1170                             raise ExtractorError(
1171                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
1172                                 expected=True)
1173                         else:
1174                             self._downloader.report_warning(
1175                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
1176                     return self.url_result('ytsearch:' + url)
1177
1178             if default_search in ('error', 'fixup_error'):
1179                 raise ExtractorError(
1180                     '%r is not a valid URL. '
1181                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
1182                     % (url, url), expected=True)
1183             else:
1184                 if ':' not in default_search:
1185                     default_search += ':'
1186                 return self.url_result(default_search + url)
1187
1188         url, smuggled_data = unsmuggle_url(url)
1189         force_videoid = None
1190         is_intentional = smuggled_data and smuggled_data.get('to_generic')
1191         if smuggled_data and 'force_videoid' in smuggled_data:
1192             force_videoid = smuggled_data['force_videoid']
1193             video_id = force_videoid
1194         else:
1195             video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
1196
1197         self.to_screen('%s: Requesting header' % video_id)
1198
1199         head_req = HEADRequest(url)
1200         head_response = self._request_webpage(
1201             head_req, video_id,
1202             note=False, errnote='Could not send HEAD request to %s' % url,
1203             fatal=False)
1204
1205         if head_response is not False:
1206             # Check for redirect
1207             new_url = head_response.geturl()
1208             if url != new_url:
1209                 self.report_following_redirect(new_url)
1210                 if force_videoid:
1211                     new_url = smuggle_url(
1212                         new_url, {'force_videoid': force_videoid})
1213                 return self.url_result(new_url)
1214
1215         full_response = None
1216         if head_response is False:
1217             request = sanitized_Request(url)
1218             request.add_header('Accept-Encoding', '*')
1219             full_response = self._request_webpage(request, video_id)
1220             head_response = full_response
1221
1222         # Check for direct link to a video
1223         content_type = head_response.headers.get('Content-Type', '')
1224         m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
1225         if m:
1226             upload_date = unified_strdate(
1227                 head_response.headers.get('Last-Modified'))
1228             return {
1229                 'id': video_id,
1230                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1231                 'direct': True,
1232                 'formats': [{
1233                     'format_id': m.group('format_id'),
1234                     'url': url,
1235                     'vcodec': 'none' if m.group('type') == 'audio' else None
1236                 }],
1237                 'upload_date': upload_date,
1238             }
1239
1240         if not self._downloader.params.get('test', False) and not is_intentional:
1241             force = self._downloader.params.get('force_generic_extractor', False)
1242             self._downloader.report_warning(
1243                 '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
1244
1245         if not full_response:
1246             request = sanitized_Request(url)
1247             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
1248             # making it impossible to download only chunk of the file (yet we need only 512kB to
1249             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
1250             # that will always result in downloading the whole file that is not desirable.
1251             # Therefore for extraction pass we have to override Accept-Encoding to any in order
1252             # to accept raw bytes and being able to download only a chunk.
1253             # It may probably better to solve this by checking Content-Type for application/octet-stream
1254             # after HEAD request finishes, but not sure if we can rely on this.
1255             request.add_header('Accept-Encoding', '*')
1256             full_response = self._request_webpage(request, video_id)
1257
1258         # Maybe it's a direct link to a video?
1259         # Be careful not to download the whole thing!
1260         first_bytes = full_response.read(512)
1261         if not is_html(first_bytes):
1262             self._downloader.report_warning(
1263                 'URL could be a direct video link, returning it as such.')
1264             upload_date = unified_strdate(
1265                 head_response.headers.get('Last-Modified'))
1266             return {
1267                 'id': video_id,
1268                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1269                 'direct': True,
1270                 'url': url,
1271                 'upload_date': upload_date,
1272             }
1273
1274         webpage = self._webpage_read_content(
1275             full_response, url, video_id, prefix=first_bytes)
1276
1277         self.report_extraction(video_id)
1278
1279         # Is it an RSS feed, a SMIL file or a XSPF playlist?
1280         try:
1281             doc = compat_etree_fromstring(webpage.encode('utf-8'))
1282             if doc.tag == 'rss':
1283                 return self._extract_rss(url, video_id, doc)
1284             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
1285                 return self._parse_smil(doc, url, video_id)
1286             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
1287                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
1288         except compat_xml_parse_error:
1289             pass
1290
1291         # Is it a Camtasia project?
1292         camtasia_res = self._extract_camtasia(url, video_id, webpage)
1293         if camtasia_res is not None:
1294             return camtasia_res
1295
1296         # Sometimes embedded video player is hidden behind percent encoding
1297         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1298         # Unescaping the whole page allows to handle those cases in a generic way
1299         webpage = compat_urllib_parse_unquote(webpage)
1300
1301         # it's tempting to parse this further, but you would
1302         # have to take into account all the variations like
1303         #   Video Title - Site Name
1304         #   Site Name | Video Title
1305         #   Video Title - Tagline | Site Name
1306         # and so on and so forth; it's just not practical
1307         video_title = self._html_search_regex(
1308             r'(?s)<title>(.*?)</title>', webpage, 'video title',
1309             default='video')
1310
1311         # Try to detect age limit automatically
1312         age_limit = self._rta_search(webpage)
1313         # And then there are the jokers who advertise that they use RTA,
1314         # but actually don't.
1315         AGE_LIMIT_MARKERS = [
1316             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1317         ]
1318         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1319             age_limit = 18
1320
1321         # video uploader is domain name
1322         video_uploader = self._search_regex(
1323             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1324
1325         # Helper method
1326         def _playlist_from_matches(matches, getter=None, ie=None):
1327             urlrs = orderedSet(
1328                 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1329                 for m in matches)
1330             return self.playlist_result(
1331                 urlrs, playlist_id=video_id, playlist_title=video_title)
1332
1333         # Look for Brightcove Legacy Studio embeds
1334         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
1335         if bc_urls:
1336             self.to_screen('Brightcove video detected.')
1337             entries = [{
1338                 '_type': 'url',
1339                 'url': smuggle_url(bc_url, {'Referer': url}),
1340                 'ie_key': 'BrightcoveLegacy'
1341             } for bc_url in bc_urls]
1342
1343             return {
1344                 '_type': 'playlist',
1345                 'title': video_title,
1346                 'id': video_id,
1347                 'entries': entries,
1348             }
1349
1350         # Look for Brightcove New Studio embeds
1351         bc_urls = BrightcoveNewIE._extract_urls(webpage)
1352         if bc_urls:
1353             return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
1354
1355         # Look for embedded rtl.nl player
1356         matches = re.findall(
1357             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1358             webpage)
1359         if matches:
1360             return _playlist_from_matches(matches, ie='RtlNl')
1361
1362         vimeo_url = VimeoIE._extract_vimeo_url(url, webpage)
1363         if vimeo_url is not None:
1364             return self.url_result(vimeo_url)
1365
1366         vid_me_embed_url = self._search_regex(
1367             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
1368             webpage, 'vid.me embed', default=None)
1369         if vid_me_embed_url is not None:
1370             return self.url_result(vid_me_embed_url, 'Vidme')
1371
1372         # Look for embedded YouTube player
1373         matches = re.findall(r'''(?x)
1374             (?:
1375                 <iframe[^>]+?src=|
1376                 data-video-url=|
1377                 <embed[^>]+?src=|
1378                 embedSWF\(?:\s*|
1379                 new\s+SWFObject\(
1380             )
1381             (["\'])
1382                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1383                 (?:embed|v|p)/.+?)
1384             \1''', webpage)
1385         if matches:
1386             return _playlist_from_matches(
1387                 matches, lambda m: unescapeHTML(m[1]))
1388
1389         # Look for lazyYT YouTube embed
1390         matches = re.findall(
1391             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1392         if matches:
1393             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1394
1395         # Look for embedded Dailymotion player
1396         matches = re.findall(
1397             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
1398         if matches:
1399             return _playlist_from_matches(
1400                 matches, lambda m: unescapeHTML(m[1]))
1401
1402         # Look for embedded Dailymotion playlist player (#3822)
1403         m = re.search(
1404             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1405         if m:
1406             playlists = re.findall(
1407                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1408             if playlists:
1409                 return _playlist_from_matches(
1410                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1411
1412         # Look for embedded Wistia player
1413         match = re.search(
1414             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1415         if match:
1416             embed_url = self._proto_relative_url(
1417                 unescapeHTML(match.group('url')))
1418             return {
1419                 '_type': 'url_transparent',
1420                 'url': embed_url,
1421                 'ie_key': 'Wistia',
1422                 'uploader': video_uploader,
1423                 'title': video_title,
1424                 'id': video_id,
1425             }
1426
1427         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1428         if match:
1429             return {
1430                 '_type': 'url_transparent',
1431                 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1432                 'ie_key': 'Wistia',
1433                 'uploader': video_uploader,
1434                 'title': video_title,
1435                 'id': match.group('id')
1436             }
1437
1438         # Look for SVT player
1439         svt_url = SVTIE._extract_url(webpage)
1440         if svt_url:
1441             return self.url_result(svt_url, 'SVT')
1442
1443         # Look for embedded condenast player
1444         matches = re.findall(
1445             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1446             webpage)
1447         if matches:
1448             return {
1449                 '_type': 'playlist',
1450                 'entries': [{
1451                     '_type': 'url',
1452                     'ie_key': 'CondeNast',
1453                     'url': ma,
1454                 } for ma in matches],
1455                 'title': video_title,
1456                 'id': video_id,
1457             }
1458
1459         # Look for Bandcamp pages with custom domain
1460         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1461         if mobj is not None:
1462             burl = unescapeHTML(mobj.group(1))
1463             # Don't set the extractor because it can be a track url or an album
1464             return self.url_result(burl)
1465
1466         # Look for embedded Vevo player
1467         mobj = re.search(
1468             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1469         if mobj is not None:
1470             return self.url_result(mobj.group('url'))
1471
1472         # Look for embedded Viddler player
1473         mobj = re.search(
1474             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1475             webpage)
1476         if mobj is not None:
1477             return self.url_result(mobj.group('url'))
1478
1479         # Look for NYTimes player
1480         mobj = re.search(
1481             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1482             webpage)
1483         if mobj is not None:
1484             return self.url_result(mobj.group('url'))
1485
1486         # Look for Libsyn player
1487         mobj = re.search(
1488             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1489         if mobj is not None:
1490             return self.url_result(mobj.group('url'))
1491
1492         # Look for Ooyala videos
1493         mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1494                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1495                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1496                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1497         if mobj is not None:
1498             return OoyalaIE._build_url_result(mobj.group('ec'))
1499
1500         # Look for multiple Ooyala embeds on SBN network websites
1501         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1502         if mobj is not None:
1503             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1504             if embeds:
1505                 return _playlist_from_matches(
1506                     embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
1507
1508         # Look for Aparat videos
1509         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1510         if mobj is not None:
1511             return self.url_result(mobj.group(1), 'Aparat')
1512
1513         # Look for MPORA videos
1514         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1515         if mobj is not None:
1516             return self.url_result(mobj.group(1), 'Mpora')
1517
1518         # Look for embedded NovaMov-based player
1519         mobj = re.search(
1520             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1521                     (?P<url>http://(?:(?:embed|www)\.)?
1522                         (?:novamov\.com|
1523                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
1524                            videoweed\.(?:es|com)|
1525                            movshare\.(?:net|sx|ag)|
1526                            divxstage\.(?:eu|net|ch|co|at|ag))
1527                         /embed\.php.+?)\1''', webpage)
1528         if mobj is not None:
1529             return self.url_result(mobj.group('url'))
1530
1531         # Look for embedded Facebook player
1532         mobj = re.search(
1533             r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1534         if mobj is not None:
1535             return self.url_result(mobj.group('url'), 'Facebook')
1536
1537         # Look for embedded VK player
1538         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1539         if mobj is not None:
1540             return self.url_result(mobj.group('url'), 'VK')
1541
1542         # Look for embedded ivi player
1543         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1544         if mobj is not None:
1545             return self.url_result(mobj.group('url'), 'Ivi')
1546
1547         # Look for embedded Huffington Post player
1548         mobj = re.search(
1549             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1550         if mobj is not None:
1551             return self.url_result(mobj.group('url'), 'HuffPost')
1552
1553         # Look for embed.ly
1554         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1555         if mobj is not None:
1556             return self.url_result(mobj.group('url'))
1557         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1558         if mobj is not None:
1559             return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
1560
1561         # Look for funnyordie embed
1562         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1563         if matches:
1564             return _playlist_from_matches(
1565                 matches, getter=unescapeHTML, ie='FunnyOrDie')
1566
1567         # Look for BBC iPlayer embed
1568         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1569         if matches:
1570             return _playlist_from_matches(matches, ie='BBCCoUk')
1571
1572         # Look for embedded RUTV player
1573         rutv_url = RUTVIE._extract_url(webpage)
1574         if rutv_url:
1575             return self.url_result(rutv_url, 'RUTV')
1576
1577         # Look for embedded TVC player
1578         tvc_url = TVCIE._extract_url(webpage)
1579         if tvc_url:
1580             return self.url_result(tvc_url, 'TVC')
1581
1582         # Look for embedded SportBox player
1583         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1584         if sportbox_urls:
1585             return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1586
1587         # Look for embedded PornHub player
1588         pornhub_url = PornHubIE._extract_url(webpage)
1589         if pornhub_url:
1590             return self.url_result(pornhub_url, 'PornHub')
1591
1592         # Look for embedded XHamster player
1593         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
1594         if xhamster_urls:
1595             return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
1596
1597         # Look for embedded Tvigle player
1598         mobj = re.search(
1599             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
1600         if mobj is not None:
1601             return self.url_result(mobj.group('url'), 'Tvigle')
1602
1603         # Look for embedded TED player
1604         mobj = re.search(
1605             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1606         if mobj is not None:
1607             return self.url_result(mobj.group('url'), 'TED')
1608
1609         # Look for embedded Ustream videos
1610         mobj = re.search(
1611             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1612         if mobj is not None:
1613             return self.url_result(mobj.group('url'), 'Ustream')
1614
1615         # Look for embedded arte.tv player
1616         mobj = re.search(
1617             r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1618             webpage)
1619         if mobj is not None:
1620             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1621
1622         # Look for embedded francetv player
1623         mobj = re.search(
1624             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
1625             webpage)
1626         if mobj is not None:
1627             return self.url_result(mobj.group('url'))
1628
1629         # Look for embedded smotri.com player
1630         smotri_url = SmotriIE._extract_url(webpage)
1631         if smotri_url:
1632             return self.url_result(smotri_url, 'Smotri')
1633
1634         # Look for embedded Myvi.ru player
1635         myvi_url = MyviIE._extract_url(webpage)
1636         if myvi_url:
1637             return self.url_result(myvi_url)
1638
1639         # Look for embeded soundcloud player
1640         mobj = re.search(
1641             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1642             webpage)
1643         if mobj is not None:
1644             url = unescapeHTML(mobj.group('url'))
1645             return self.url_result(url)
1646
1647         # Look for embedded vulture.com player
1648         mobj = re.search(
1649             r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1650             webpage)
1651         if mobj is not None:
1652             url = unescapeHTML(mobj.group('url'))
1653             return self.url_result(url, ie='Vulture')
1654
1655         # Look for embedded mtvservices player
1656         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
1657         if mtvservices_url:
1658             return self.url_result(mtvservices_url, ie='MTVServicesEmbedded')
1659
1660         # Look for embedded yahoo player
1661         mobj = re.search(
1662             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1663             webpage)
1664         if mobj is not None:
1665             return self.url_result(mobj.group('url'), 'Yahoo')
1666
1667         # Look for embedded sbs.com.au player
1668         mobj = re.search(
1669             r'''(?x)
1670             (?:
1671                 <meta\s+property="og:video"\s+content=|
1672                 <iframe[^>]+?src=
1673             )
1674             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1675             webpage)
1676         if mobj is not None:
1677             return self.url_result(mobj.group('url'), 'SBS')
1678
1679         # Look for embedded Cinchcast player
1680         mobj = re.search(
1681             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1682             webpage)
1683         if mobj is not None:
1684             return self.url_result(mobj.group('url'), 'Cinchcast')
1685
1686         mobj = re.search(
1687             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1688             webpage)
1689         if not mobj:
1690             mobj = re.search(
1691                 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
1692                 webpage)
1693         if mobj is not None:
1694             return self.url_result(mobj.group('url'), 'MLB')
1695
1696         mobj = re.search(
1697             r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1698             webpage)
1699         if mobj is not None:
1700             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1701
1702         mobj = re.search(
1703             r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1704             webpage)
1705         if mobj is not None:
1706             return self.url_result(mobj.group('url'), 'Livestream')
1707
1708         # Look for Zapiks embed
1709         mobj = re.search(
1710             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1711         if mobj is not None:
1712             return self.url_result(mobj.group('url'), 'Zapiks')
1713
1714         # Look for Kaltura embeds
1715         mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or
1716                 re.search(r'(?s)(?P<q1>["\'])(?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?(?P=q1).*?entry_?[Ii]d\s*:\s*(?P<q2>["\'])(?P<id>.+?)(?P=q2)', webpage))
1717         if mobj is not None:
1718             return self.url_result(smuggle_url(
1719                 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(),
1720                 {'source_url': url}), 'Kaltura')
1721
1722         # Look for Eagle.Platform embeds
1723         mobj = re.search(
1724             r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1725         if mobj is not None:
1726             return self.url_result(mobj.group('url'), 'EaglePlatform')
1727
1728         # Look for ClipYou (uses Eagle.Platform) embeds
1729         mobj = re.search(
1730             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1731         if mobj is not None:
1732             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1733
1734         # Look for Pladform embeds
1735         mobj = re.search(
1736             r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
1737         if mobj is not None:
1738             return self.url_result(mobj.group('url'), 'Pladform')
1739
1740         # Look for Playwire embeds
1741         mobj = re.search(
1742             r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
1743         if mobj is not None:
1744             return self.url_result(mobj.group('url'))
1745
1746         # Look for 5min embeds
1747         mobj = re.search(
1748             r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1749         if mobj is not None:
1750             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1751
1752         # Look for Crooks and Liars embeds
1753         mobj = re.search(
1754             r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1755         if mobj is not None:
1756             return self.url_result(mobj.group('url'))
1757
1758         # Look for NBC Sports VPlayer embeds
1759         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1760         if nbc_sports_url:
1761             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1762
1763         # Look for UDN embeds
1764         mobj = re.search(
1765             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage)
1766         if mobj is not None:
1767             return self.url_result(
1768                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1769
1770         # Look for Senate ISVP iframe
1771         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
1772         if senate_isvp_url:
1773             return self.url_result(senate_isvp_url, 'SenateISVP')
1774
1775         # Look for Dailymotion Cloud videos
1776         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
1777         if dmcloud_url:
1778             return self.url_result(dmcloud_url, 'DailymotionCloud')
1779
1780         # Look for OnionStudios embeds
1781         onionstudios_url = OnionStudiosIE._extract_url(webpage)
1782         if onionstudios_url:
1783             return self.url_result(onionstudios_url)
1784
1785         # Look for SnagFilms embeds
1786         snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage)
1787         if snagfilms_url:
1788             return self.url_result(snagfilms_url)
1789
1790         # Look for ScreenwaveMedia embeds
1791         mobj = re.search(ScreenwaveMediaIE.EMBED_PATTERN, webpage)
1792         if mobj is not None:
1793             return self.url_result(unescapeHTML(mobj.group('url')), 'ScreenwaveMedia')
1794
1795         # Look for AdobeTVVideo embeds
1796         mobj = re.search(
1797             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
1798             webpage)
1799         if mobj is not None:
1800             return self.url_result(
1801                 self._proto_relative_url(unescapeHTML(mobj.group(1))),
1802                 'AdobeTVVideo')
1803
1804         def check_video(vurl):
1805             if YoutubeIE.suitable(vurl):
1806                 return True
1807             vpath = compat_urlparse.urlparse(vurl).path
1808             vext = determine_ext(vpath)
1809             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1810
1811         def filter_video(urls):
1812             return list(filter(check_video, urls))
1813
1814         # Start with something easy: JW Player in SWFObject
1815         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1816         if not found:
1817             # Look for gorilla-vid style embedding
1818             found = filter_video(re.findall(r'''(?sx)
1819                 (?:
1820                     jw_plugins|
1821                     JWPlayerOptions|
1822                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1823                 )
1824                 .*?
1825                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1826         if not found:
1827             # Broaden the search a little bit
1828             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1829         if not found:
1830             # Broaden the findall a little bit: JWPlayer JS loader
1831             found = filter_video(re.findall(
1832                 r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1833         if not found:
1834             # Flow player
1835             found = filter_video(re.findall(r'''(?xs)
1836                 flowplayer\("[^"]+",\s*
1837                     \{[^}]+?\}\s*,
1838                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1839                         ["']?url["']?\s*:\s*["']([^"']+)["']
1840             ''', webpage))
1841         if not found:
1842             # Cinerama player
1843             found = re.findall(
1844                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1845         if not found:
1846             # Try to find twitter cards info
1847             found = filter_video(re.findall(
1848                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1849         if not found:
1850             # We look for Open Graph info:
1851             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1852             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1853             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1854             if m_video_type is not None:
1855                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1856         if not found:
1857             # HTML5 video
1858             found = re.findall(r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1859         if not found:
1860             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1861             found = re.search(
1862                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1863                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1864                 webpage)
1865             if not found:
1866                 # Look also in Refresh HTTP header
1867                 refresh_header = head_response.headers.get('Refresh')
1868                 if refresh_header:
1869                     # In python 2 response HTTP headers are bytestrings
1870                     if sys.version_info < (3, 0) and isinstance(refresh_header, str):
1871                         refresh_header = refresh_header.decode('iso-8859-1')
1872                     found = re.search(REDIRECT_REGEX, refresh_header)
1873             if found:
1874                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
1875                 self.report_following_redirect(new_url)
1876                 return {
1877                     '_type': 'url',
1878                     'url': new_url,
1879                 }
1880         if not found:
1881             raise UnsupportedError(url)
1882
1883         entries = []
1884         for video_url in found:
1885             video_url = video_url.replace('\\/', '/')
1886             video_url = compat_urlparse.urljoin(url, video_url)
1887             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
1888
1889             # Sometimes, jwplayer extraction will result in a YouTube URL
1890             if YoutubeIE.suitable(video_url):
1891                 entries.append(self.url_result(video_url, 'Youtube'))
1892                 continue
1893
1894             # here's a fun little line of code for you:
1895             video_id = os.path.splitext(video_id)[0]
1896
1897             entry_info_dict = {
1898                 'id': video_id,
1899                 'uploader': video_uploader,
1900                 'title': video_title,
1901                 'age_limit': age_limit,
1902             }
1903
1904             ext = determine_ext(video_url)
1905             if ext == 'smil':
1906                 entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
1907             elif ext == 'xspf':
1908                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
1909             elif ext == 'm3u8':
1910                 entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
1911             else:
1912                 entry_info_dict['url'] = video_url
1913
1914             entries.append(entry_info_dict)
1915
1916         if len(entries) == 1:
1917             return entries[0]
1918         else:
1919             for num, e in enumerate(entries, start=1):
1920                 # 'url' results don't have a title
1921                 if e.get('title') is not None:
1922                     e['title'] = '%s (%d)' % (e['title'], num)
1923             return {
1924                 '_type': 'playlist',
1925                 'entries': entries,
1926             }