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

   1 # coding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 import os
   6 import re
   7 import sys
   8
   9 from .common import InfoExtractor
  10 from .youtube import YoutubeIE
  11 from ..compat import (
  12     compat_etree_fromstring,
  13     compat_urllib_parse_unquote,
  14     compat_urlparse,
  15     compat_xml_parse_error,
  16 )
  17 from ..utils import (
  18     determine_ext,
  19     ExtractorError,
  20     float_or_none,
  21     HEADRequest,
  22     is_html,
  23     js_to_json,
  24     orderedSet,
  25     sanitized_Request,
  26     smuggle_url,
  27     unescapeHTML,
  28     unified_strdate,
  29     unsmuggle_url,
  30     UnsupportedError,
  31     xpath_text,
  32 )
  33 from .commonprotocols import RtmpIE
  34 from .brightcove import (
  35     BrightcoveLegacyIE,
  36     BrightcoveNewIE,
  37 )
  38 from .nbc import NBCSportsVPlayerIE
  39 from .ooyala import OoyalaIE
  40 from .rutv import RUTVIE
  41 from .tvc import TVCIE
  42 from .sportbox import SportBoxEmbedIE
  43 from .smotri import SmotriIE
  44 from .myvi import MyviIE
  45 from .condenast import CondeNastIE
  46 from .udn import UDNEmbedIE
  47 from .senateisvp import SenateISVPIE
  48 from .svt import SVTIE
  49 from .pornhub import PornHubIE
  50 from .xhamster import XHamsterEmbedIE
  51 from .tnaflix import TNAFlixNetworkEmbedIE
  52 from .drtuber import DrTuberIE
  53 from .redtube import RedTubeIE
  54 from .vimeo import VimeoIE
  55 from .dailymotion import (
  56     DailymotionIE,
  57     DailymotionCloudIE,
  58 )
  59 from .onionstudios import OnionStudiosIE
  60 from .viewlift import ViewLiftEmbedIE
  61 from .mtv import MTVServicesEmbeddedIE
  62 from .pladform import PladformIE
  63 from .videomore import VideomoreIE
  64 from .webcaster import WebcasterFeedIE
  65 from .googledrive import GoogleDriveIE
  66 from .jwplatform import JWPlatformIE
  67 from .digiteka import DigitekaIE
  68 from .arkena import ArkenaIE
  69 from .instagram import InstagramIE
  70 from .liveleak import LiveLeakIE
  71 from .threeqsdn import ThreeQSDNIE
  72 from .theplatform import ThePlatformIE
  73 from .vessel import VesselIE
  74 from .kaltura import KalturaIE
  75 from .eagleplatform import EaglePlatformIE
  76 from .facebook import FacebookIE
  77 from .soundcloud import SoundcloudIE
  78 from .tunein import TuneInBaseIE
  79 from .vbox7 import Vbox7IE
  80 from .dbtv import DBTVIE
  81 from .piksel import PikselIE
  82 from .videa import VideaIE
  83 from .twentymin import TwentyMinutenIE
  84 from .ustream import UstreamIE
  85 from .openload import OpenloadIE
  86 from .videopress import VideoPressIE
  87 from .rutube import RutubeIE
  88 from .limelight import LimelightBaseIE
  89
  90
  91 class GenericIE(InfoExtractor):
  92     IE_DESC = 'Generic downloader that works on some sites'
  93     _VALID_URL = r'.*'
  94     IE_NAME = 'generic'
  95     _TESTS = [
  96         # Direct link to a video
  97         {
  98             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  99             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
 100             'info_dict': {
 101                 'id': 'trailer',
 102                 'ext': 'mp4',
 103                 'title': 'trailer',
 104                 'upload_date': '20100513',
 105             }
 106         },
 107         # Direct link to media delivered compressed (until Accept-Encoding is *)
 108         {
 109             'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
 110             'md5': '128c42e68b13950268b648275386fc74',
 111             'info_dict': {
 112                 'id': 'FictionJunction-Parallel_Hearts',
 113                 'ext': 'flac',
 114                 'title': 'FictionJunction-Parallel_Hearts',
 115                 'upload_date': '20140522',
 116             },
 117             'expected_warnings': [
 118                 'URL could be a direct video link, returning it as such.'
 119             ],
 120             'skip': 'URL invalid',
 121         },
 122         # Direct download with broken HEAD
 123         {
 124             'url': 'http://ai-radio.org:8000/radio.opus',
 125             'info_dict': {
 126                 'id': 'radio',
 127                 'ext': 'opus',
 128                 'title': 'radio',
 129             },
 130             'params': {
 131                 'skip_download': True,  # infinite live stream
 132             },
 133             'expected_warnings': [
 134                 r'501.*Not Implemented',
 135                 r'400.*Bad Request',
 136             ],
 137         },
 138         # Direct link with incorrect MIME type
 139         {
 140             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 141             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 142             'info_dict': {
 143                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 144                 'id': '5_Lennart_Poettering_-_Systemd',
 145                 'ext': 'webm',
 146                 'title': '5_Lennart_Poettering_-_Systemd',
 147                 'upload_date': '20141120',
 148             },
 149             'expected_warnings': [
 150                 'URL could be a direct video link, returning it as such.'
 151             ]
 152         },
 153         # RSS feed
 154         {
 155             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 156             'info_dict': {
 157                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 158                 'title': 'Zero Punctuation',
 159                 'description': 're:.*groundbreaking video review series.*'
 160             },
 161             'playlist_mincount': 11,
 162         },
 163         # RSS feed with enclosure
 164         {
 165             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 166             'info_dict': {
 167                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 168                 'ext': 'm4v',
 169                 'upload_date': '20150228',
 170                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 171             }
 172         },
 173         # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
 174         {
 175             'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
 176             'info_dict': {
 177                 'id': 'smil',
 178                 'ext': 'mp4',
 179                 'title': 'Automatics, robotics and biocybernetics',
 180                 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
 181                 'upload_date': '20130627',
 182                 'formats': 'mincount:16',
 183                 'subtitles': 'mincount:1',
 184             },
 185             'params': {
 186                 'force_generic_extractor': True,
 187                 'skip_download': True,
 188             },
 189         },
 190         # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html
 191         {
 192             'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil',
 193             'info_dict': {
 194                 'id': 'hds',
 195                 'ext': 'flv',
 196                 'title': 'hds',
 197                 'formats': 'mincount:1',
 198             },
 199             'params': {
 200                 'skip_download': True,
 201             },
 202         },
 203         # SMIL from https://www.restudy.dk/video/play/id/1637
 204         {
 205             'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml',
 206             'info_dict': {
 207                 'id': 'video_1637',
 208                 'ext': 'flv',
 209                 'title': 'video_1637',
 210                 'formats': 'mincount:3',
 211             },
 212             'params': {
 213                 'skip_download': True,
 214             },
 215         },
 216         # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm
 217         {
 218             'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil',
 219             'info_dict': {
 220                 'id': 'smil-service',
 221                 'ext': 'flv',
 222                 'title': 'smil-service',
 223                 'formats': 'mincount:1',
 224             },
 225             'params': {
 226                 'skip_download': True,
 227             },
 228         },
 229         # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370
 230         {
 231             'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil',
 232             'info_dict': {
 233                 'id': '4719370',
 234                 'ext': 'mp4',
 235                 'title': '571de1fd-47bc-48db-abf9-238872a58d1f',
 236                 'formats': 'mincount:3',
 237             },
 238             'params': {
 239                 'skip_download': True,
 240             },
 241         },
 242         # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html
 243         {
 244             'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf',
 245             'info_dict': {
 246                 'id': 'mZlp2ctYIUEB',
 247                 'ext': 'mp4',
 248                 'title': 'Tikibad ontruimd wegens brand',
 249                 'description': 'md5:05ca046ff47b931f9b04855015e163a4',
 250                 'thumbnail': r're:^https?://.*\.jpg$',
 251                 'duration': 33,
 252             },
 253             'params': {
 254                 'skip_download': True,
 255             },
 256         },
 257         # MPD from http://dash-mse-test.appspot.com/media.html
 258         {
 259             'url': 'http://yt-dash-mse-test.commondatastorage.googleapis.com/media/car-20120827-manifest.mpd',
 260             'md5': '4b57baab2e30d6eb3a6a09f0ba57ef53',
 261             'info_dict': {
 262                 'id': 'car-20120827-manifest',
 263                 'ext': 'mp4',
 264                 'title': 'car-20120827-manifest',
 265                 'formats': 'mincount:9',
 266                 'upload_date': '20130904',
 267             },
 268             'params': {
 269                 'format': 'bestvideo',
 270             },
 271         },
 272         # m3u8 served with Content-Type: audio/x-mpegURL; charset=utf-8
 273         {
 274             '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',
 275             'info_dict': {
 276                 'id': 'content',
 277                 'ext': 'mp4',
 278                 'title': 'content',
 279                 'formats': 'mincount:8',
 280             },
 281             'params': {
 282                 # m3u8 downloads
 283                 'skip_download': True,
 284             },
 285             'skip': 'video gone',
 286         },
 287         # m3u8 served with Content-Type: text/plain
 288         {
 289             'url': 'http://www.nacentapps.com/m3u8/index.m3u8',
 290             'info_dict': {
 291                 'id': 'index',
 292                 'ext': 'mp4',
 293                 'title': 'index',
 294                 'upload_date': '20140720',
 295                 'formats': 'mincount:11',
 296             },
 297             'params': {
 298                 # m3u8 downloads
 299                 'skip_download': True,
 300             },
 301             'skip': 'video gone',
 302         },
 303         # google redirect
 304         {
 305             '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',
 306             'info_dict': {
 307                 'id': 'cmQHVoWB5FY',
 308                 'ext': 'mp4',
 309                 'upload_date': '20130224',
 310                 'uploader_id': 'TheVerge',
 311                 'description': r're:^Chris Ziegler takes a look at the\.*',
 312                 'uploader': 'The Verge',
 313                 'title': 'First Firefox OS phones side-by-side',
 314             },
 315             'params': {
 316                 'skip_download': False,
 317             }
 318         },
 319         {
 320             # redirect in Refresh HTTP header
 321             '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',
 322             'info_dict': {
 323                 'id': 'pO8h3EaFRdo',
 324                 'ext': 'mp4',
 325                 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
 326                 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5',
 327                 'upload_date': '20150917',
 328                 'uploader_id': 'brtvofficial',
 329                 'uploader': 'Boiler Room',
 330             },
 331             'params': {
 332                 'skip_download': False,
 333             },
 334         },
 335         {
 336             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
 337             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
 338             'info_dict': {
 339                 'id': '13601338388002',
 340                 'ext': 'mp4',
 341                 'uploader': 'www.hodiho.fr',
 342                 'title': 'R\u00e9gis plante sa Jeep',
 343             }
 344         },
 345         # bandcamp page with custom domain
 346         {
 347             'add_ie': ['Bandcamp'],
 348             'url': 'http://bronyrock.com/track/the-pony-mash',
 349             'info_dict': {
 350                 'id': '3235767654',
 351                 'ext': 'mp3',
 352                 'title': 'The Pony Mash',
 353                 'uploader': 'M_Pallante',
 354             },
 355             'skip': 'There is a limit of 200 free downloads / month for the test song',
 356         },
 357         {
 358             # embedded brightcove video
 359             # it also tests brightcove videos that need to set the 'Referer'
 360             # in the http requests
 361             'add_ie': ['BrightcoveLegacy'],
 362             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
 363             'info_dict': {
 364                 'id': '2765128793001',
 365                 'ext': 'mp4',
 366                 'title': 'Le cours de bourse : l’analyse technique',
 367                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
 368                 'uploader': 'BFM BUSINESS',
 369             },
 370             'params': {
 371                 'skip_download': True,
 372             },
 373         },
 374         {
 375             # embedded with itemprop embedURL and video id spelled as `idVideo`
 376             'add_id': ['BrightcoveLegacy'],
 377             'url': 'http://bfmbusiness.bfmtv.com/mediaplayer/chroniques/olivier-delamarche/',
 378             'info_dict': {
 379                 'id': '5255628253001',
 380                 'ext': 'mp4',
 381                 'title': 'md5:37c519b1128915607601e75a87995fc0',
 382                 'description': 'md5:37f7f888b434bb8f8cc8dbd4f7a4cf26',
 383                 'uploader': 'BFM BUSINESS',
 384                 'uploader_id': '876450612001',
 385                 'timestamp': 1482255315,
 386                 'upload_date': '20161220',
 387             },
 388             'params': {
 389                 'skip_download': True,
 390             },
 391         },
 392         {
 393             # https://github.com/rg3/youtube-dl/issues/2253
 394             'url': 'http://bcove.me/i6nfkrc3',
 395             'md5': '0ba9446db037002366bab3b3eb30c88c',
 396             'info_dict': {
 397                 'id': '3101154703001',
 398                 'ext': 'mp4',
 399                 'title': 'Still no power',
 400                 'uploader': 'thestar.com',
 401                 '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.',
 402             },
 403             'add_ie': ['BrightcoveLegacy'],
 404             'skip': 'video gone',
 405         },
 406         {
 407             'url': 'http://www.championat.com/video/football/v/87/87499.html',
 408             'md5': 'fb973ecf6e4a78a67453647444222983',
 409             'info_dict': {
 410                 'id': '3414141473001',
 411                 'ext': 'mp4',
 412                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 413                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 414                 'uploader': 'Championat',
 415             },
 416         },
 417         {
 418             # https://github.com/rg3/youtube-dl/issues/3541
 419             'add_ie': ['BrightcoveLegacy'],
 420             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 421             'info_dict': {
 422                 'id': '3866516442001',
 423                 'ext': 'mp4',
 424                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 425                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 426                 'uploader': 'SBS Broadcasting',
 427             },
 428             'skip': 'Restricted to Netherlands',
 429             'params': {
 430                 'skip_download': True,  # m3u8 download
 431             },
 432         },
 433         {
 434             # Brightcove video in <iframe>
 435             'url': 'http://www.un.org/chinese/News/story.asp?NewsID=27724',
 436             'md5': '36d74ef5e37c8b4a2ce92880d208b968',
 437             'info_dict': {
 438                 'id': '5360463607001',
 439                 'ext': 'mp4',
 440                 'title': '叙利亚失明儿童在废墟上演唱《心跳》  呼吁获得正常童年生活',
 441                 'description': '联合国儿童基金会中东和北非区域大使、作曲家扎德·迪拉尼（Zade Dirani）在3月15日叙利亚冲突爆发7周年纪念日之际发布了为叙利亚谱写的歌曲《心跳》（HEARTBEAT），为受到六年冲突影响的叙利亚儿童发出强烈呐喊，呼吁世界做出共同努力，使叙利亚儿童重新获得享有正常童年生活的权利。',
 442                 'uploader': 'United Nations',
 443                 'uploader_id': '1362235914001',
 444                 'timestamp': 1489593889,
 445                 'upload_date': '20170315',
 446             },
 447             'add_ie': ['BrightcoveLegacy'],
 448         },
 449         {
 450             # Brightcove with alternative playerID key
 451             'url': 'http://www.nature.com/nmeth/journal/v9/n7/fig_tab/nmeth.2062_SV1.html',
 452             'info_dict': {
 453                 'id': 'nmeth.2062_SV1',
 454                 'title': 'Simultaneous multiview imaging of the Drosophila syncytial blastoderm : Quantitative high-speed imaging of entire developing embryos with simultaneous multiview light-sheet microscopy : Nature Methods : Nature Research',
 455             },
 456             'playlist': [{
 457                 'info_dict': {
 458                     'id': '2228375078001',
 459                     'ext': 'mp4',
 460                     'title': 'nmeth.2062-sv1',
 461                     'description': 'nmeth.2062-sv1',
 462                     'timestamp': 1363357591,
 463                     'upload_date': '20130315',
 464                     'uploader': 'Nature Publishing Group',
 465                     'uploader_id': '1964492299001',
 466                 },
 467             }],
 468         },
 469         {
 470             # Brightcove with UUID in videoPlayer
 471             'url': 'http://www8.hp.com/cn/zh/home.html',
 472             'info_dict': {
 473                 'id': '5255815316001',
 474                 'ext': 'mp4',
 475                 'title': 'Sprocket Video - China',
 476                 'description': 'Sprocket Video - China',
 477                 'uploader': 'HP-Video Gallery',
 478                 'timestamp': 1482263210,
 479                 'upload_date': '20161220',
 480                 'uploader_id': '1107601872001',
 481             },
 482             'params': {
 483                 'skip_download': True,  # m3u8 download
 484             },
 485             'skip': 'video rotates...weekly?',
 486         },
 487         {
 488             # Brightcove:new type [2].
 489             'url': 'http://www.delawaresportszone.com/video-st-thomas-more-earns-first-trip-to-basketball-semis',
 490             'md5': '2b35148fcf48da41c9fb4591650784f3',
 491             'info_dict': {
 492                 'id': '5348741021001',
 493                 'ext': 'mp4',
 494                 'upload_date': '20170306',
 495                 'uploader_id': '4191638492001',
 496                 'timestamp': 1488769918,
 497                 'title': 'VIDEO:  St. Thomas More earns first trip to basketball semis',
 498
 499             },
 500         },
 501         {
 502             # Alternative brightcove <video> attributes
 503             'url': 'http://www.programme-tv.net/videos/extraits/81095-guillaume-canet-evoque-les-rumeurs-d-infidelite-de-marion-cotillard-avec-brad-pitt-dans-vivement-dimanche/',
 504             'info_dict': {
 505                 'id': '81095-guillaume-canet-evoque-les-rumeurs-d-infidelite-de-marion-cotillard-avec-brad-pitt-dans-vivement-dimanche',
 506                 'title': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche, Extraits : toutes les vidéos avec Télé-Loisirs",
 507             },
 508             'playlist': [{
 509                 'md5': '732d22ba3d33f2f3fc253c39f8f36523',
 510                 'info_dict': {
 511                     'id': '5311302538001',
 512                     'ext': 'mp4',
 513                     'title': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche",
 514                     'description': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche (France 2, 5 février 2017)",
 515                     'timestamp': 1486321708,
 516                     'upload_date': '20170205',
 517                     'uploader_id': '800000640001',
 518                 },
 519                 'only_matching': True,
 520             }],
 521         },
 522         {
 523             # Brightcove with UUID in videoPlayer
 524             'url': 'http://www8.hp.com/cn/zh/home.html',
 525             'info_dict': {
 526                 'id': '5255815316001',
 527                 'ext': 'mp4',
 528                 'title': 'Sprocket Video - China',
 529                 'description': 'Sprocket Video - China',
 530                 'uploader': 'HP-Video Gallery',
 531                 'timestamp': 1482263210,
 532                 'upload_date': '20161220',
 533                 'uploader_id': '1107601872001',
 534             },
 535             'params': {
 536                 'skip_download': True,  # m3u8 download
 537             },
 538         },
 539         # ooyala video
 540         {
 541             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 542             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 543             'info_dict': {
 544                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 545                 'ext': 'mp4',
 546                 'title': '2cc213299525360.mov',  # that's what we get
 547                 'duration': 238.231,
 548             },
 549             'add_ie': ['Ooyala'],
 550         },
 551         {
 552             # ooyala video embedded with http://player.ooyala.com/iframe.js
 553             'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/',
 554             'info_dict': {
 555                 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB',
 556                 'ext': 'mp4',
 557                 'title': '"Steve Jobs: Man in the Machine" trailer',
 558                 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
 559                 'duration': 135.427,
 560             },
 561             'params': {
 562                 'skip_download': True,
 563             },
 564             'skip': 'movie expired',
 565         },
 566         # embed.ly video
 567         {
 568             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 569             'info_dict': {
 570                 'id': '9ODmcdjQcHQ',
 571                 'ext': 'mp4',
 572                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 573                 'upload_date': '20140225',
 574                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 575                 'uploader': 'Tested',
 576                 'uploader_id': 'testedcom',
 577             },
 578             # No need to test YoutubeIE here
 579             'params': {
 580                 'skip_download': True,
 581             },
 582         },
 583         # funnyordie embed
 584         {
 585             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 586             'info_dict': {
 587                 'id': '18e820ec3f',
 588                 'ext': 'mp4',
 589                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 590                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 591             },
 592             # HEAD requests lead to endless 301, while GET is OK
 593             'expected_warnings': ['301'],
 594         },
 595         # RUTV embed
 596         {
 597             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 598             'info_dict': {
 599                 'id': '776940',
 600                 'ext': 'mp4',
 601                 'title': 'Охотское море стало целиком российским',
 602                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 603             },
 604             'params': {
 605                 # m3u8 download
 606                 'skip_download': True,
 607             },
 608         },
 609         # TVC embed
 610         {
 611             '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/',
 612             'info_dict': {
 613                 'id': '55304',
 614                 'ext': 'mp4',
 615                 'title': 'Дошкольное воспитание',
 616             },
 617         },
 618         # SportBox embed
 619         {
 620             'url': 'http://www.vestifinance.ru/articles/25753',
 621             'info_dict': {
 622                 'id': '25753',
 623                 'title': 'Прямые трансляции с Форума-выставки "Госзаказ-2013"',
 624             },
 625             'playlist': [{
 626                 'info_dict': {
 627                     'id': '370908',
 628                     'title': 'Госзаказ. День 3',
 629                     'ext': 'mp4',
 630                 }
 631             }, {
 632                 'info_dict': {
 633                     'id': '370905',
 634                     'title': 'Госзаказ. День 2',
 635                     'ext': 'mp4',
 636                 }
 637             }, {
 638                 'info_dict': {
 639                     'id': '370902',
 640                     'title': 'Госзаказ. День 1',
 641                     'ext': 'mp4',
 642                 }
 643             }],
 644             'params': {
 645                 # m3u8 download
 646                 'skip_download': True,
 647             },
 648         },
 649         # Myvi.ru embed
 650         {
 651             'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
 652             'info_dict': {
 653                 'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
 654                 'ext': 'mp4',
 655                 'title': 'Ужастики, русский трейлер (2015)',
 656                 'thumbnail': r're:^https?://.*\.jpg$',
 657                 'duration': 153,
 658             }
 659         },
 660         # XHamster embed
 661         {
 662             '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',
 663             'info_dict': {
 664                 'id': 'showthread',
 665                 'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
 666             },
 667             'playlist_mincount': 7,
 668             # This forum does not allow <iframe> syntaxes anymore
 669             # Now HTML tags are displayed as-is
 670             'skip': 'No videos on this page',
 671         },
 672         # Embedded TED video
 673         {
 674             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 675             'md5': '65fdff94098e4a607385a60c5177c638',
 676             'info_dict': {
 677                 'id': '1969',
 678                 'ext': 'mp4',
 679                 'title': 'Hidden miracles of the natural world',
 680                 'uploader': 'Louie Schwartzberg',
 681                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 682             }
 683         },
 684         # nowvideo embed hidden behind percent encoding
 685         {
 686             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 687             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 688             'info_dict': {
 689                 'id': '06e53103ca9aa',
 690                 'ext': 'flv',
 691                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 692                 'description': 'No description',
 693             },
 694         },
 695         # arte embed
 696         {
 697             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 698             'md5': '7653032cbb25bf6c80d80f217055fa43',
 699             'info_dict': {
 700                 'id': '048195-004_PLUS7-F',
 701                 'ext': 'flv',
 702                 'title': 'X:enius',
 703                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 704                 'upload_date': '20140320',
 705             },
 706             'params': {
 707                 'skip_download': 'Requires rtmpdump'
 708             },
 709             'skip': 'video gone',
 710         },
 711         # francetv embed
 712         {
 713             'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
 714             'info_dict': {
 715                 'id': 'EV_30231',
 716                 'ext': 'mp4',
 717                 'title': 'Alcaline, le concert avec Calogero',
 718                 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
 719                 'upload_date': '20150226',
 720                 'timestamp': 1424989860,
 721                 'duration': 5400,
 722             },
 723             'params': {
 724                 # m3u8 downloads
 725                 'skip_download': True,
 726             },
 727             'expected_warnings': [
 728                 'Forbidden'
 729             ]
 730         },
 731         # Condé Nast embed
 732         {
 733             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 734             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 735             'info_dict': {
 736                 'id': '53501be369702d3275860000',
 737                 'ext': 'mp4',
 738                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 739             }
 740         },
 741         # Dailymotion embed
 742         {
 743             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 744             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 745             'info_dict': {
 746                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 747                 'ext': 'mp4',
 748                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 749                 'description': 'md5:faf028e48a461b8b7fad38f1e104b119',
 750                 'uploader': 'Spi0n',
 751                 'uploader_id': 'xgditw',
 752                 'upload_date': '20140425',
 753                 'timestamp': 1398441542,
 754             },
 755             'add_ie': ['Dailymotion'],
 756         },
 757         # YouTube embed
 758         {
 759             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 760             'info_dict': {
 761                 'id': 'FXRb4ykk4S0',
 762                 'ext': 'mp4',
 763                 'title': 'The NBL Auction 2014',
 764                 'uploader': 'BADMINTON England',
 765                 'uploader_id': 'BADMINTONEvents',
 766                 'upload_date': '20140603',
 767                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 768             },
 769             'add_ie': ['Youtube'],
 770             'params': {
 771                 'skip_download': True,
 772             }
 773         },
 774         # MTVSercices embed
 775         {
 776             'url': 'http://www.vulture.com/2016/06/new-key-peele-sketches-released.html',
 777             'md5': 'ca1aef97695ef2c1d6973256a57e5252',
 778             'info_dict': {
 779                 'id': '769f7ec0-0692-4d62-9b45-0d88074bffc1',
 780                 'ext': 'mp4',
 781                 'title': 'Key and Peele|October 10, 2012|2|203|Liam Neesons - Uncensored',
 782                 'description': 'Two valets share their love for movie star Liam Neesons.',
 783                 'timestamp': 1349922600,
 784                 'upload_date': '20121011',
 785             },
 786         },
 787         # YouTube embed via <data-embed-url="">
 788         {
 789             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 790             'info_dict': {
 791                 'id': '4vAffPZIT44',
 792                 'ext': 'mp4',
 793                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 794                 'uploader': 'Gameloft',
 795                 'uploader_id': 'gameloft',
 796                 'upload_date': '20140828',
 797                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 798             },
 799             'params': {
 800                 'skip_download': True,
 801             }
 802         },
 803         # YouTube <object> embed
 804         {
 805             'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
 806             'md5': '516718101ec834f74318df76259fb3cc',
 807             'info_dict': {
 808                 'id': 'msN87y-iEx0',
 809                 'ext': 'webm',
 810                 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
 811                 'upload_date': '20080526',
 812                 'description': 'md5:0ffc78ea3f01b2e2c247d5f8d1d3c18d',
 813                 'uploader': 'Christopher Sykes',
 814                 'uploader_id': 'ChristopherJSykes',
 815             },
 816             'add_ie': ['Youtube'],
 817         },
 818         # Camtasia studio
 819         {
 820             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 821             'playlist': [{
 822                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 823                 'info_dict': {
 824                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 825                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 826                     'ext': 'flv',
 827                     'duration': 2235.90,
 828                 }
 829             }, {
 830                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 831                 'info_dict': {
 832                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 833                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 834                     'ext': 'flv',
 835                     'duration': 2235.93,
 836                 }
 837             }],
 838             'info_dict': {
 839                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 840             }
 841         },
 842         # Flowplayer
 843         {
 844             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 845             'md5': '9d65602bf31c6e20014319c7d07fba27',
 846             'info_dict': {
 847                 'id': '5123ea6d5e5a7',
 848                 'ext': 'mp4',
 849                 'age_limit': 18,
 850                 'uploader': 'www.handjobhub.com',
 851                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 852             }
 853         },
 854         # Multiple brightcove videos
 855         # https://github.com/rg3/youtube-dl/issues/2283
 856         {
 857             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 858             'info_dict': {
 859                 'id': 'always-never',
 860                 'title': 'Always / Never - The New Yorker',
 861             },
 862             'playlist_count': 3,
 863             'params': {
 864                 'extract_flat': False,
 865                 'skip_download': True,
 866             }
 867         },
 868         # MLB embed
 869         {
 870             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 871             'md5': '96f09a37e44da40dd083e12d9a683327',
 872             'info_dict': {
 873                 'id': '33322633',
 874                 'ext': 'mp4',
 875                 'title': 'Ump changes call to ball',
 876                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 877                 'duration': 48,
 878                 'timestamp': 1401537900,
 879                 'upload_date': '20140531',
 880                 'thumbnail': r're:^https?://.*\.jpg$',
 881             },
 882         },
 883         # Wistia embed
 884         {
 885             'url': 'http://study.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 886             'md5': '1953f3a698ab51cfc948ed3992a0b7ff',
 887             'info_dict': {
 888                 'id': '6e2wtrbdaf',
 889                 'ext': 'mov',
 890                 'title': 'paywall_north-american-exploration-failed-colonies-of-spain-france-england',
 891                 'description': 'a Paywall Videos video from Remilon',
 892                 'duration': 644.072,
 893                 'uploader': 'study.com',
 894                 'timestamp': 1459678540,
 895                 'upload_date': '20160403',
 896                 'filesize': 24687186,
 897             },
 898         },
 899         {
 900             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 901             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 902             'info_dict': {
 903                 'id': 'uxjb0lwrcz',
 904                 'ext': 'mp4',
 905                 'title': 'Conversation about Hexagonal Rails Part 1',
 906                 'description': 'a Martin Fowler video from ThoughtWorks',
 907                 'duration': 1715.0,
 908                 'uploader': 'thoughtworks.wistia.com',
 909                 'timestamp': 1401832161,
 910                 'upload_date': '20140603',
 911             },
 912         },
 913         # Wistia standard embed (async)
 914         {
 915             'url': 'https://www.getdrip.com/university/brennan-dunn-drip-workshop/',
 916             'info_dict': {
 917                 'id': '807fafadvk',
 918                 'ext': 'mp4',
 919                 'title': 'Drip Brennan Dunn Workshop',
 920                 'description': 'a JV Webinars video from getdrip-1',
 921                 'duration': 4986.95,
 922                 'timestamp': 1463607249,
 923                 'upload_date': '20160518',
 924             },
 925             'params': {
 926                 'skip_download': True,
 927             }
 928         },
 929         # Soundcloud embed
 930         {
 931             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 932             'info_dict': {
 933                 'id': '174391317',
 934                 'ext': 'mp3',
 935                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 936                 'uploader': 'Sophos Security',
 937                 'title': 'Chet Chat 171 - Oct 29, 2014',
 938                 'upload_date': '20141029',
 939             }
 940         },
 941         # Soundcloud multiple embeds
 942         {
 943             'url': 'http://www.guitarplayer.com/lessons/1014/legato-workout-one-hour-to-more-fluid-performance---tab/52809',
 944             'info_dict': {
 945                 'id': '52809',
 946                 'title': 'Guitar Essentials: Legato Workout—One-Hour to Fluid Performance  | TAB + AUDIO',
 947             },
 948             'playlist_mincount': 7,
 949         },
 950         # TuneIn station embed
 951         {
 952             'url': 'http://radiocnrv.com/promouvoir-radio-cnrv/',
 953             'info_dict': {
 954                 'id': '204146',
 955                 'ext': 'mp3',
 956                 'title': 'CNRV',
 957                 'location': 'Paris, France',
 958                 'is_live': True,
 959             },
 960             'params': {
 961                 # Live stream
 962                 'skip_download': True,
 963             },
 964         },
 965         # Livestream embed
 966         {
 967             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 968             'info_dict': {
 969                 'id': '67864563',
 970                 'ext': 'flv',
 971                 'upload_date': '20141112',
 972                 'title': 'Rosetta #CometLanding webcast HL 10',
 973             }
 974         },
 975         # Another Livestream embed, without 'new.' in URL
 976         {
 977             'url': 'https://www.freespeech.org/',
 978             'info_dict': {
 979                 'id': '123537347',
 980                 'ext': 'mp4',
 981                 'title': 're:^FSTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 982             },
 983             'params': {
 984                 # Live stream
 985                 'skip_download': True,
 986             },
 987         },
 988         # LazyYT
 989         {
 990             'url': 'https://skiplagged.com/',
 991             'info_dict': {
 992                 'id': 'skiplagged',
 993                 'title': 'Skiplagged: The smart way to find cheap flights',
 994             },
 995             'playlist_mincount': 1,
 996             'add_ie': ['Youtube'],
 997         },
 998         # Cinchcast embed
 999         {
1000             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
1001             'info_dict': {
1002                 'id': '7141703',
1003                 'ext': 'mp3',
1004                 'upload_date': '20141126',
1005                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
1006             }
1007         },
1008         # Cinerama player
1009         {
1010             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
1011             'info_dict': {
1012                 'id': '730m_DandD_1901_512k',
1013                 'ext': 'mp4',
1014                 'uploader': 'www.abc.net.au',
1015                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
1016             }
1017         },
1018         # embedded viddler video
1019         {
1020             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
1021             'info_dict': {
1022                 'id': '4d03aad9',
1023                 'ext': 'mp4',
1024                 'uploader': 'deadspin',
1025                 'title': 'WALL-TO-GORTAT',
1026                 'timestamp': 1422285291,
1027                 'upload_date': '20150126',
1028             },
1029             'add_ie': ['Viddler'],
1030         },
1031         # Libsyn embed
1032         {
1033             'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
1034             'info_dict': {
1035                 'id': '3377616',
1036                 'ext': 'mp3',
1037                 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
1038                 'description': 'md5:601cb790edd05908957dae8aaa866465',
1039                 'upload_date': '20150220',
1040             },
1041             'skip': 'All The Daily Show URLs now redirect to http://www.cc.com/shows/',
1042         },
1043         # jwplayer YouTube
1044         {
1045             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
1046             'info_dict': {
1047                 'id': 'Mrj4DVp2zeA',
1048                 'ext': 'mp4',
1049                 'upload_date': '20150212',
1050                 'uploader': 'The National Archives UK',
1051                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
1052                 'uploader_id': 'NationalArchives08',
1053                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
1054             },
1055         },
1056         # jwplayer rtmp
1057         {
1058             'url': 'http://www.suffolk.edu/sjc/',
1059             'info_dict': {
1060                 'id': 'sjclive',
1061                 'ext': 'flv',
1062                 'title': 'Massachusetts Supreme Judicial Court Oral Arguments',
1063                 'uploader': 'www.suffolk.edu',
1064             },
1065             'params': {
1066                 'skip_download': True,
1067             }
1068         },
1069         # Complex jwplayer
1070         {
1071             'url': 'http://www.indiedb.com/games/king-machine/videos',
1072             'info_dict': {
1073                 'id': 'videos',
1074                 'ext': 'mp4',
1075                 'title': 'king machine trailer 1',
1076                 'thumbnail': r're:^https?://.*\.jpg$',
1077             },
1078         },
1079         {
1080             # JWPlayer config passed as variable
1081             'url': 'http://www.txxx.com/videos/3326530/ariele/',
1082             'info_dict': {
1083                 'id': '3326530_hq',
1084                 'ext': 'mp4',
1085                 'title': 'ARIELE | Tube Cup',
1086                 'uploader': 'www.txxx.com',
1087                 'age_limit': 18,
1088             },
1089             'params': {
1090                 'skip_download': True,
1091             }
1092         },
1093         # rtl.nl embed
1094         {
1095             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
1096             'playlist_mincount': 5,
1097             'info_dict': {
1098                 'id': 'aanslagen-kopenhagen',
1099                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
1100             }
1101         },
1102         # Zapiks embed
1103         {
1104             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
1105             'info_dict': {
1106                 'id': '118046',
1107                 'ext': 'mp4',
1108                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
1109             }
1110         },
1111         # Kaltura embed (different embed code)
1112         {
1113             'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
1114             'info_dict': {
1115                 'id': '1_a52wc67y',
1116                 'ext': 'flv',
1117                 'upload_date': '20150127',
1118                 'uploader_id': 'PremierMedia',
1119                 'timestamp': int,
1120                 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
1121             },
1122         },
1123         # Kaltura embed with single quotes
1124         {
1125             'url': 'http://fod.infobase.com/p_ViewPlaylist.aspx?AssignmentID=NUN8ZY',
1126             'info_dict': {
1127                 'id': '0_izeg5utt',
1128                 'ext': 'mp4',
1129                 'title': '35871',
1130                 'timestamp': 1355743100,
1131                 'upload_date': '20121217',
1132                 'uploader_id': 'batchUser',
1133             },
1134             'add_ie': ['Kaltura'],
1135         },
1136         {
1137             # Kaltura embedded via quoted entry_id
1138             'url': 'https://www.oreilly.com/ideas/my-cloud-makes-pretty-pictures',
1139             'info_dict': {
1140                 'id': '0_utuok90b',
1141                 'ext': 'mp4',
1142                 'title': '06_matthew_brender_raj_dutt',
1143                 'timestamp': 1466638791,
1144                 'upload_date': '20160622',
1145             },
1146             'add_ie': ['Kaltura'],
1147             'expected_warnings': [
1148                 'Could not send HEAD request'
1149             ],
1150             'params': {
1151                 'skip_download': True,
1152             }
1153         },
1154         {
1155             # Kaltura embedded, some fileExt broken (#11480)
1156             'url': 'http://www.cornell.edu/video/nima-arkani-hamed-standard-models-of-particle-physics',
1157             'info_dict': {
1158                 'id': '1_sgtvehim',
1159                 'ext': 'mp4',
1160                 'title': 'Our "Standard Models" of particle physics and cosmology',
1161                 'description': 'md5:67ea74807b8c4fea92a6f38d6d323861',
1162                 'timestamp': 1321158993,
1163                 'upload_date': '20111113',
1164                 'uploader_id': 'kps1',
1165             },
1166             'add_ie': ['Kaltura'],
1167         },
1168         {
1169             # Kaltura iframe embed
1170             'url': 'http://www.gsd.harvard.edu/event/i-m-pei-a-centennial-celebration/',
1171             'md5': 'ae5ace8eb09dc1a35d03b579a9c2cc44',
1172             'info_dict': {
1173                 'id': '0_f2cfbpwy',
1174                 'ext': 'mp4',
1175                 'title': 'I. M. Pei: A Centennial Celebration',
1176                 'description': 'md5:1db8f40c69edc46ca180ba30c567f37c',
1177                 'upload_date': '20170403',
1178                 'uploader_id': 'batchUser',
1179                 'timestamp': 1491232186,
1180             },
1181             'add_ie': ['Kaltura'],
1182         },
1183         # Eagle.Platform embed (generic URL)
1184         {
1185             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
1186             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
1187             'info_dict': {
1188                 'id': '227304',
1189                 'ext': 'mp4',
1190                 'title': 'Навальный вышел на свободу',
1191                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
1192                 'thumbnail': r're:^https?://.*\.jpg$',
1193                 'duration': 87,
1194                 'view_count': int,
1195                 'age_limit': 0,
1196             },
1197         },
1198         # ClipYou (Eagle.Platform) embed (custom URL)
1199         {
1200             'url': 'http://muz-tv.ru/play/7129/',
1201             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
1202             'info_dict': {
1203                 'id': '12820',
1204                 'ext': 'mp4',
1205                 'title': "'O Sole Mio",
1206                 'thumbnail': r're:^https?://.*\.jpg$',
1207                 'duration': 216,
1208                 'view_count': int,
1209             },
1210         },
1211         # Pladform embed
1212         {
1213             'url': 'http://muz-tv.ru/kinozal/view/7400/',
1214             'info_dict': {
1215                 'id': '100183293',
1216                 'ext': 'mp4',
1217                 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
1218                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
1219                 'thumbnail': r're:^https?://.*\.jpg$',
1220                 'duration': 694,
1221                 'age_limit': 0,
1222             },
1223         },
1224         # Playwire embed
1225         {
1226             'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
1227             'info_dict': {
1228                 'id': '3519514',
1229                 'ext': 'mp4',
1230                 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
1231                 'thumbnail': r're:^https?://.*\.png$',
1232                 'duration': 45.115,
1233             },
1234         },
1235         # 5min embed
1236         {
1237             'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
1238             'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
1239             'info_dict': {
1240                 'id': '518726732',
1241                 'ext': 'mp4',
1242                 'title': 'Facebook Creates "On This Day" | Crunch Report',
1243             },
1244         },
1245         # SVT embed
1246         {
1247             'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
1248             'info_dict': {
1249                 'id': '2900353',
1250                 'ext': 'flv',
1251                 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
1252                 'duration': 27,
1253                 'age_limit': 0,
1254             },
1255         },
1256         # Crooks and Liars embed
1257         {
1258             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
1259             'info_dict': {
1260                 'id': '8RUoRhRi',
1261                 'ext': 'mp4',
1262                 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
1263                 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
1264                 'timestamp': 1428207000,
1265                 'upload_date': '20150405',
1266                 'uploader': 'Heather',
1267             },
1268         },
1269         # Crooks and Liars external embed
1270         {
1271             'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
1272             'info_dict': {
1273                 'id': 'MTE3MjUtMzQ2MzA',
1274                 'ext': 'mp4',
1275                 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
1276                 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
1277                 'timestamp': 1265032391,
1278                 'upload_date': '20100201',
1279                 'uploader': 'Heather',
1280             },
1281         },
1282         # NBC Sports vplayer embed
1283         {
1284             'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
1285             'info_dict': {
1286                 'id': 'ln7x1qSThw4k',
1287                 'ext': 'flv',
1288                 'title': "PFT Live: New leader in the 'new-look' defense",
1289                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
1290                 'uploader': 'NBCU-SPORTS',
1291                 'upload_date': '20140107',
1292                 'timestamp': 1389118457,
1293             },
1294         },
1295         # NBC News embed
1296         {
1297             'url': 'http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html',
1298             'md5': '1aa589c675898ae6d37a17913cf68d66',
1299             'info_dict': {
1300                 'id': '701714499682',
1301                 'ext': 'mp4',
1302                 'title': 'PREVIEW: On Assignment: David Letterman',
1303                 '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.',
1304             },
1305         },
1306         # UDN embed
1307         {
1308             'url': 'https://video.udn.com/news/300346',
1309             'md5': 'fd2060e988c326991037b9aff9df21a6',
1310             'info_dict': {
1311                 'id': '300346',
1312                 'ext': 'mp4',
1313                 'title': '中一中男師變性 全校師生力挺',
1314                 'thumbnail': r're:^https?://.*\.jpg$',
1315             },
1316             'params': {
1317                 # m3u8 download
1318                 'skip_download': True,
1319             },
1320         },
1321         # Ooyala embed
1322         {
1323             'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
1324             'info_dict': {
1325                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
1326                 'ext': 'mp4',
1327                 'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.',
1328                 'title': 'This is what separates the Excel masters from the wannabes',
1329                 'duration': 191.933,
1330             },
1331             'params': {
1332                 # m3u8 downloads
1333                 'skip_download': True,
1334             }
1335         },
1336         # Brightcove URL in single quotes
1337         {
1338             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
1339             'md5': '4ae374f1f8b91c889c4b9203c8c752af',
1340             'info_dict': {
1341                 'id': '4255764656001',
1342                 'ext': 'mp4',
1343                 'title': 'SN Presents: Russell Martin, World Citizen',
1344                 '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.',
1345                 'uploader': 'Rogers Sportsnet',
1346                 'uploader_id': '1704050871',
1347                 'upload_date': '20150525',
1348                 'timestamp': 1432570283,
1349             },
1350         },
1351         # Dailymotion Cloud video
1352         {
1353             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
1354             'md5': 'dcaf23ad0c67a256f4278bce6e0bae38',
1355             'info_dict': {
1356                 'id': 'x2uy8t3',
1357                 'ext': 'mp4',
1358                 'title': 'Sauvons les abeilles ! - Le débat',
1359                 'description': 'md5:d9082128b1c5277987825d684939ca26',
1360                 'thumbnail': r're:^https?://.*\.jpe?g$',
1361                 'timestamp': 1434970506,
1362                 'upload_date': '20150622',
1363                 'uploader': 'Public Sénat',
1364                 'uploader_id': 'xa9gza',
1365             }
1366         },
1367         # OnionStudios embed
1368         {
1369             'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
1370             'info_dict': {
1371                 'id': '2855',
1372                 'ext': 'mp4',
1373                 'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
1374                 'thumbnail': r're:^https?://.*\.jpe?g$',
1375                 'uploader': 'ClickHole',
1376                 'uploader_id': 'clickhole',
1377             }
1378         },
1379         # SnagFilms embed
1380         {
1381             'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
1382             'info_dict': {
1383                 'id': '74849a00-85a9-11e1-9660-123139220831',
1384                 'ext': 'mp4',
1385                 'title': '#whilewewatch',
1386             }
1387         },
1388         # AdobeTVVideo embed
1389         {
1390             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
1391             'md5': '43662b577c018ad707a63766462b1e87',
1392             'info_dict': {
1393                 'id': '2456',
1394                 'ext': 'mp4',
1395                 'title': 'New experience with Acrobat DC',
1396                 'description': 'New experience with Acrobat DC',
1397                 'duration': 248.667,
1398             },
1399         },
1400         # BrightcoveInPageEmbed embed
1401         {
1402             'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
1403             'info_dict': {
1404                 'id': '4238694884001',
1405                 'ext': 'flv',
1406                 'title': 'Tabletop: Dread, Last Thoughts',
1407                 'description': 'Tabletop: Dread, Last Thoughts',
1408                 'duration': 51690,
1409             },
1410         },
1411         # Brightcove embed, with no valid 'renditions' but valid 'IOSRenditions'
1412         # This video can't be played in browsers if Flash disabled and UA set to iPhone, which is actually a false alarm
1413         {
1414             'url': 'https://dl.dropboxusercontent.com/u/29092637/interview.html',
1415             'info_dict': {
1416                 'id': '4785848093001',
1417                 'ext': 'mp4',
1418                 'title': 'The Cardinal Pell Interview',
1419                 'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ',
1420                 'uploader': 'GlobeCast Australia - GlobeStream',
1421                 'uploader_id': '2733773828001',
1422                 'upload_date': '20160304',
1423                 'timestamp': 1457083087,
1424             },
1425             'params': {
1426                 # m3u8 downloads
1427                 'skip_download': True,
1428             },
1429         },
1430         # Another form of arte.tv embed
1431         {
1432             'url': 'http://www.tv-replay.fr/redirection/09-04-16/arte-reportage-arte-11508975.html',
1433             'md5': '850bfe45417ddf221288c88a0cffe2e2',
1434             'info_dict': {
1435                 'id': '030273-562_PLUS7-F',
1436                 'ext': 'mp4',
1437                 'title': 'ARTE Reportage - Nulle part, en France',
1438                 'description': 'md5:e3a0e8868ed7303ed509b9e3af2b870d',
1439                 'upload_date': '20160409',
1440             },
1441         },
1442         # LiveLeak embed
1443         {
1444             'url': 'http://www.wykop.pl/link/3088787/',
1445             'md5': 'ace83b9ed19b21f68e1b50e844fdf95d',
1446             'info_dict': {
1447                 'id': '874_1459135191',
1448                 'ext': 'mp4',
1449                 'title': 'Man shows poor quality of new apartment building',
1450                 'description': 'The wall is like a sand pile.',
1451                 'uploader': 'Lake8737',
1452             }
1453         },
1454         # Duplicated embedded video URLs
1455         {
1456             'url': 'http://www.hudl.com/athlete/2538180/highlights/149298443',
1457             'info_dict': {
1458                 'id': '149298443_480_16c25b74_2',
1459                 'ext': 'mp4',
1460                 'title': 'vs. Blue Orange Spring Game',
1461                 'uploader': 'www.hudl.com',
1462             },
1463         },
1464         # twitter:player:stream embed
1465         {
1466             'url': 'http://www.rtl.be/info/video/589263.aspx?CategoryID=288',
1467             'info_dict': {
1468                 'id': 'master',
1469                 'ext': 'mp4',
1470                 'title': 'Une nouvelle espèce de dinosaure découverte en Argentine',
1471                 'uploader': 'www.rtl.be',
1472             },
1473             'params': {
1474                 # m3u8 downloads
1475                 'skip_download': True,
1476             },
1477         },
1478         # twitter:player embed
1479         {
1480             'url': 'http://www.theatlantic.com/video/index/484130/what-do-black-holes-sound-like/',
1481             'md5': 'a3e0df96369831de324f0778e126653c',
1482             'info_dict': {
1483                 'id': '4909620399001',
1484                 'ext': 'mp4',
1485                 'title': 'What Do Black Holes Sound Like?',
1486                 'description': 'what do black holes sound like',
1487                 'upload_date': '20160524',
1488                 'uploader_id': '29913724001',
1489                 'timestamp': 1464107587,
1490                 'uploader': 'TheAtlantic',
1491             },
1492             'add_ie': ['BrightcoveLegacy'],
1493         },
1494         # Facebook <iframe> embed
1495         {
1496             'url': 'https://www.hostblogger.de/blog/archives/6181-Auto-jagt-Betonmischer.html',
1497             'md5': 'fbcde74f534176ecb015849146dd3aee',
1498             'info_dict': {
1499                 'id': '599637780109885',
1500                 'ext': 'mp4',
1501                 'title': 'Facebook video #599637780109885',
1502             },
1503         },
1504         # Facebook API embed
1505         {
1506             'url': 'http://www.lothype.com/blue-stars-2016-preview-standstill-full-show/',
1507             'md5': 'a47372ee61b39a7b90287094d447d94e',
1508             'info_dict': {
1509                 'id': '10153467542406923',
1510                 'ext': 'mp4',
1511                 'title': 'Facebook video #10153467542406923',
1512             },
1513         },
1514         # Wordpress "YouTube Video Importer" plugin
1515         {
1516             'url': 'http://www.lothype.com/blue-devils-drumline-stanford-lot-2016/',
1517             'md5': 'd16797741b560b485194eddda8121b48',
1518             'info_dict': {
1519                 'id': 'HNTXWDXV9Is',
1520                 'ext': 'mp4',
1521                 'title': 'Blue Devils Drumline Stanford lot 2016',
1522                 'upload_date': '20160627',
1523                 'uploader_id': 'GENOCIDE8GENERAL10',
1524                 'uploader': 'cylus cyrus',
1525             },
1526         },
1527         {
1528             # video stored on custom kaltura server
1529             'url': 'http://www.expansion.com/multimedia/videos.html?media=EQcM30NHIPv',
1530             'md5': '537617d06e64dfed891fa1593c4b30cc',
1531             'info_dict': {
1532                 'id': '0_1iotm5bh',
1533                 'ext': 'mp4',
1534                 'title': 'Elecciones británicas: 5 lecciones para Rajoy',
1535                 'description': 'md5:435a89d68b9760b92ce67ed227055f16',
1536                 'uploader_id': 'videos.expansion@el-mundo.net',
1537                 'upload_date': '20150429',
1538                 'timestamp': 1430303472,
1539             },
1540             'add_ie': ['Kaltura'],
1541         },
1542         {
1543             # Non-standard Vimeo embed
1544             'url': 'https://openclassrooms.com/courses/understanding-the-web',
1545             'md5': '64d86f1c7d369afd9a78b38cbb88d80a',
1546             'info_dict': {
1547                 'id': '148867247',
1548                 'ext': 'mp4',
1549                 'title': 'Understanding the web - Teaser',
1550                 'description': 'This is "Understanding the web - Teaser" by openclassrooms on Vimeo, the home for high quality videos and the people who love them.',
1551                 'upload_date': '20151214',
1552                 'uploader': 'OpenClassrooms',
1553                 'uploader_id': 'openclassrooms',
1554             },
1555             'add_ie': ['Vimeo'],
1556         },
1557         {
1558             # generic vimeo embed that requires original URL passed as Referer
1559             'url': 'http://racing4everyone.eu/2016/07/30/formula-1-2016-round12-germany/',
1560             'only_matching': True,
1561         },
1562         {
1563             'url': 'https://support.arkena.com/display/PLAY/Ways+to+embed+your+video',
1564             'md5': 'b96f2f71b359a8ecd05ce4e1daa72365',
1565             'info_dict': {
1566                 'id': 'b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe',
1567                 'ext': 'mp4',
1568                 'title': 'Big Buck Bunny',
1569                 'description': 'Royalty free test video',
1570                 'timestamp': 1432816365,
1571                 'upload_date': '20150528',
1572                 'is_live': False,
1573             },
1574             'params': {
1575                 'skip_download': True,
1576             },
1577             'add_ie': [ArkenaIE.ie_key()],
1578         },
1579         {
1580             '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/',
1581             'info_dict': {
1582                 'id': '1c7141f46c',
1583                 'ext': 'mp4',
1584                 'title': 'НА КОСЪМ ОТ ВЗРИВ: Изтичане на газ на бензиностанция в Пловдив',
1585             },
1586             'params': {
1587                 'skip_download': True,
1588             },
1589             'add_ie': [Vbox7IE.ie_key()],
1590         },
1591         {
1592             # DBTV embeds
1593             'url': 'http://www.dagbladet.no/2016/02/23/nyheter/nordlys/ski/troms/ver/43254897/',
1594             'info_dict': {
1595                 'id': '43254897',
1596                 'title': 'Etter ett års planlegging, klaffet endelig alt: - Jeg måtte ta en liten dans',
1597             },
1598             'playlist_mincount': 3,
1599         },
1600         {
1601             # Videa embeds
1602             'url': 'http://forum.dvdtalk.com/movie-talk/623756-deleted-magic-star-wars-ot-deleted-alt-scenes-docu-style.html',
1603             'info_dict': {
1604                 'id': '623756-deleted-magic-star-wars-ot-deleted-alt-scenes-docu-style',
1605                 'title': 'Deleted Magic - Star Wars: OT Deleted / Alt. Scenes Docu. Style - DVD Talk Forum',
1606             },
1607             'playlist_mincount': 2,
1608         },
1609         {
1610             # 20 minuten embed
1611             'url': 'http://www.20min.ch/schweiz/news/story/So-kommen-Sie-bei-Eis-und-Schnee-sicher-an-27032552',
1612             'info_dict': {
1613                 'id': '523629',
1614                 'ext': 'mp4',
1615                 'title': 'So kommen Sie bei Eis und Schnee sicher an',
1616                 'description': 'md5:117c212f64b25e3d95747e5276863f7d',
1617             },
1618             'params': {
1619                 'skip_download': True,
1620             },
1621             'add_ie': [TwentyMinutenIE.ie_key()],
1622         },
1623         {
1624             # VideoPress embed
1625             'url': 'https://en.support.wordpress.com/videopress/',
1626             'info_dict': {
1627                 'id': 'OcobLTqC',
1628                 'ext': 'm4v',
1629                 'title': 'IMG_5786',
1630                 'timestamp': 1435711927,
1631                 'upload_date': '20150701',
1632             },
1633             'params': {
1634                 'skip_download': True,
1635             },
1636             'add_ie': [VideoPressIE.ie_key()],
1637         },
1638         {
1639             # Rutube embed
1640             'url': 'http://magazzino.friday.ru/videos/vipuski/kazan-2',
1641             'info_dict': {
1642                 'id': '9b3d5bee0a8740bf70dfd29d3ea43541',
1643                 'ext': 'flv',
1644                 'title': 'Магаззино: Казань 2',
1645                 'description': 'md5:99bccdfac2269f0e8fdbc4bbc9db184a',
1646                 'uploader': 'Магаззино',
1647                 'upload_date': '20170228',
1648                 'uploader_id': '996642',
1649             },
1650             'params': {
1651                 'skip_download': True,
1652             },
1653             'add_ie': [RutubeIE.ie_key()],
1654         },
1655         {
1656             # ThePlatform embedded with whitespaces in URLs
1657             'url': 'http://www.golfchannel.com/topics/shows/golftalkcentral.htm',
1658             'only_matching': True,
1659         },
1660         {
1661             # Senate ISVP iframe https
1662             'url': 'https://www.hsgac.senate.gov/hearings/canadas-fast-track-refugee-plan-unanswered-questions-and-implications-for-us-national-security',
1663             'md5': 'fb8c70b0b515e5037981a2492099aab8',
1664             'info_dict': {
1665                 'id': 'govtaff020316',
1666                 'ext': 'mp4',
1667                 'title': 'Integrated Senate Video Player',
1668             },
1669             'add_ie': [SenateISVPIE.ie_key()],
1670         },
1671         {
1672             # Limelight embeds (1 channel embed + 4 media embeds)
1673             'url': 'http://www.sedona.com/FacilitatorTraining2017',
1674             'info_dict': {
1675                 'id': 'FacilitatorTraining2017',
1676                 'title': 'Facilitator Training 2017',
1677             },
1678             'playlist_mincount': 5,
1679         },
1680         # {
1681         #     # TODO: find another test
1682         #     # http://schema.org/VideoObject
1683         #     'url': 'https://flipagram.com/f/nyvTSJMKId',
1684         #     'md5': '888dcf08b7ea671381f00fab74692755',
1685         #     'info_dict': {
1686         #         'id': 'nyvTSJMKId',
1687         #         'ext': 'mp4',
1688         #         'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction',
1689         #         'description': '#love for cats.',
1690         #         'timestamp': 1461244995,
1691         #         'upload_date': '20160421',
1692         #     },
1693         #     'params': {
1694         #         'force_generic_extractor': True,
1695         #     },
1696         # }
1697     ]
1698
1699     def report_following_redirect(self, new_url):
1700         """Report information extraction."""
1701         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
1702
1703     def _extract_rss(self, url, video_id, doc):
1704         playlist_title = doc.find('./channel/title').text
1705         playlist_desc_el = doc.find('./channel/description')
1706         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
1707
1708         entries = []
1709         for it in doc.findall('./channel/item'):
1710             next_url = xpath_text(it, 'link', fatal=False)
1711             if not next_url:
1712                 enclosure_nodes = it.findall('./enclosure')
1713                 for e in enclosure_nodes:
1714                     next_url = e.attrib.get('url')
1715                     if next_url:
1716                         break
1717
1718             if not next_url:
1719                 continue
1720
1721             entries.append({
1722                 '_type': 'url_transparent',
1723                 'url': next_url,
1724                 'title': it.find('title').text,
1725             })
1726
1727         return {
1728             '_type': 'playlist',
1729             'id': url,
1730             'title': playlist_title,
1731             'description': playlist_desc,
1732             'entries': entries,
1733         }
1734
1735     def _extract_camtasia(self, url, video_id, webpage):
1736         """ Returns None if no camtasia video can be found. """
1737
1738         camtasia_cfg = self._search_regex(
1739             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
1740             webpage, 'camtasia configuration file', default=None)
1741         if camtasia_cfg is None:
1742             return None
1743
1744         title = self._html_search_meta('DC.title', webpage, fatal=True)
1745
1746         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
1747         camtasia_cfg = self._download_xml(
1748             camtasia_url, video_id,
1749             note='Downloading camtasia configuration',
1750             errnote='Failed to download camtasia configuration')
1751         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
1752
1753         entries = []
1754         for n in fileset_node.getchildren():
1755             url_n = n.find('./uri')
1756             if url_n is None:
1757                 continue
1758
1759             entries.append({
1760                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
1761                 'title': '%s - %s' % (title, n.tag),
1762                 'url': compat_urlparse.urljoin(url, url_n.text),
1763                 'duration': float_or_none(n.find('./duration').text),
1764             })
1765
1766         return {
1767             '_type': 'playlist',
1768             'entries': entries,
1769             'title': title,
1770         }
1771
1772     def _real_extract(self, url):
1773         if url.startswith('//'):
1774             return {
1775                 '_type': 'url',
1776                 'url': self.http_scheme() + url,
1777             }
1778
1779         parsed_url = compat_urlparse.urlparse(url)
1780         if not parsed_url.scheme:
1781             default_search = self._downloader.params.get('default_search')
1782             if default_search is None:
1783                 default_search = 'fixup_error'
1784
1785             if default_search in ('auto', 'auto_warning', 'fixup_error'):
1786                 if '/' in url:
1787                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
1788                     return self.url_result('http://' + url)
1789                 elif default_search != 'fixup_error':
1790                     if default_search == 'auto_warning':
1791                         if re.match(r'^(?:url|URL)$', url):
1792                             raise ExtractorError(
1793                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
1794                                 expected=True)
1795                         else:
1796                             self._downloader.report_warning(
1797                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
1798                     return self.url_result('ytsearch:' + url)
1799
1800             if default_search in ('error', 'fixup_error'):
1801                 raise ExtractorError(
1802                     '%r is not a valid URL. '
1803                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
1804                     % (url, url), expected=True)
1805             else:
1806                 if ':' not in default_search:
1807                     default_search += ':'
1808                 return self.url_result(default_search + url)
1809
1810         url, smuggled_data = unsmuggle_url(url)
1811         force_videoid = None
1812         is_intentional = smuggled_data and smuggled_data.get('to_generic')
1813         if smuggled_data and 'force_videoid' in smuggled_data:
1814             force_videoid = smuggled_data['force_videoid']
1815             video_id = force_videoid
1816         else:
1817             video_id = self._generic_id(url)
1818
1819         self.to_screen('%s: Requesting header' % video_id)
1820
1821         head_req = HEADRequest(url)
1822         head_response = self._request_webpage(
1823             head_req, video_id,
1824             note=False, errnote='Could not send HEAD request to %s' % url,
1825             fatal=False)
1826
1827         if head_response is not False:
1828             # Check for redirect
1829             new_url = head_response.geturl()
1830             if url != new_url:
1831                 self.report_following_redirect(new_url)
1832                 if force_videoid:
1833                     new_url = smuggle_url(
1834                         new_url, {'force_videoid': force_videoid})
1835                 return self.url_result(new_url)
1836
1837         full_response = None
1838         if head_response is False:
1839             request = sanitized_Request(url)
1840             request.add_header('Accept-Encoding', '*')
1841             full_response = self._request_webpage(request, video_id)
1842             head_response = full_response
1843
1844         info_dict = {
1845             'id': video_id,
1846             'title': self._generic_title(url),
1847             'upload_date': unified_strdate(head_response.headers.get('Last-Modified'))
1848         }
1849
1850         # Check for direct link to a video
1851         content_type = head_response.headers.get('Content-Type', '').lower()
1852         m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>[^;\s]+)', content_type)
1853         if m:
1854             format_id = m.group('format_id')
1855             if format_id.endswith('mpegurl'):
1856                 formats = self._extract_m3u8_formats(url, video_id, 'mp4')
1857             elif format_id == 'f4m':
1858                 formats = self._extract_f4m_formats(url, video_id)
1859             else:
1860                 formats = [{
1861                     'format_id': m.group('format_id'),
1862                     'url': url,
1863                     'vcodec': 'none' if m.group('type') == 'audio' else None
1864                 }]
1865                 info_dict['direct'] = True
1866             self._sort_formats(formats)
1867             info_dict['formats'] = formats
1868             return info_dict
1869
1870         if not self._downloader.params.get('test', False) and not is_intentional:
1871             force = self._downloader.params.get('force_generic_extractor', False)
1872             self._downloader.report_warning(
1873                 '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
1874
1875         if not full_response:
1876             request = sanitized_Request(url)
1877             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
1878             # making it impossible to download only chunk of the file (yet we need only 512kB to
1879             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
1880             # that will always result in downloading the whole file that is not desirable.
1881             # Therefore for extraction pass we have to override Accept-Encoding to any in order
1882             # to accept raw bytes and being able to download only a chunk.
1883             # It may probably better to solve this by checking Content-Type for application/octet-stream
1884             # after HEAD request finishes, but not sure if we can rely on this.
1885             request.add_header('Accept-Encoding', '*')
1886             full_response = self._request_webpage(request, video_id)
1887
1888         first_bytes = full_response.read(512)
1889
1890         # Is it an M3U playlist?
1891         if first_bytes.startswith(b'#EXTM3U'):
1892             info_dict['formats'] = self._extract_m3u8_formats(url, video_id, 'mp4')
1893             self._sort_formats(info_dict['formats'])
1894             return info_dict
1895
1896         # Maybe it's a direct link to a video?
1897         # Be careful not to download the whole thing!
1898         if not is_html(first_bytes):
1899             self._downloader.report_warning(
1900                 'URL could be a direct video link, returning it as such.')
1901             info_dict.update({
1902                 'direct': True,
1903                 'url': url,
1904             })
1905             return info_dict
1906
1907         webpage = self._webpage_read_content(
1908             full_response, url, video_id, prefix=first_bytes)
1909
1910         self.report_extraction(video_id)
1911
1912         # Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest?
1913         try:
1914             doc = compat_etree_fromstring(webpage.encode('utf-8'))
1915             if doc.tag == 'rss':
1916                 return self._extract_rss(url, video_id, doc)
1917             elif doc.tag == 'SmoothStreamingMedia':
1918                 info_dict['formats'] = self._parse_ism_formats(doc, url)
1919                 self._sort_formats(info_dict['formats'])
1920                 return info_dict
1921             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
1922                 smil = self._parse_smil(doc, url, video_id)
1923                 self._sort_formats(smil['formats'])
1924                 return smil
1925             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
1926                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
1927             elif re.match(r'(?i)^(?:{[^}]+})?MPD$', doc.tag):
1928                 info_dict['formats'] = self._parse_mpd_formats(
1929                     doc, video_id,
1930                     mpd_base_url=full_response.geturl().rpartition('/')[0],
1931                     mpd_url=url)
1932                 self._sort_formats(info_dict['formats'])
1933                 return info_dict
1934             elif re.match(r'^{http://ns\.adobe\.com/f4m/[12]\.0}manifest$', doc.tag):
1935                 info_dict['formats'] = self._parse_f4m_formats(doc, url, video_id)
1936                 self._sort_formats(info_dict['formats'])
1937                 return info_dict
1938         except compat_xml_parse_error:
1939             pass
1940
1941         # Is it a Camtasia project?
1942         camtasia_res = self._extract_camtasia(url, video_id, webpage)
1943         if camtasia_res is not None:
1944             return camtasia_res
1945
1946         # Sometimes embedded video player is hidden behind percent encoding
1947         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1948         # Unescaping the whole page allows to handle those cases in a generic way
1949         webpage = compat_urllib_parse_unquote(webpage)
1950
1951         # it's tempting to parse this further, but you would
1952         # have to take into account all the variations like
1953         #   Video Title - Site Name
1954         #   Site Name | Video Title
1955         #   Video Title - Tagline | Site Name
1956         # and so on and so forth; it's just not practical
1957         video_title = self._og_search_title(
1958             webpage, default=None) or self._html_search_regex(
1959             r'(?s)<title>(.*?)</title>', webpage, 'video title',
1960             default='video')
1961
1962         # Try to detect age limit automatically
1963         age_limit = self._rta_search(webpage)
1964         # And then there are the jokers who advertise that they use RTA,
1965         # but actually don't.
1966         AGE_LIMIT_MARKERS = [
1967             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1968         ]
1969         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1970             age_limit = 18
1971
1972         # video uploader is domain name
1973         video_uploader = self._search_regex(
1974             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1975
1976         video_description = self._og_search_description(webpage, default=None)
1977         video_thumbnail = self._og_search_thumbnail(webpage, default=None)
1978
1979         # Look for Brightcove Legacy Studio embeds
1980         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
1981         if bc_urls:
1982             entries = [{
1983                 '_type': 'url',
1984                 'url': smuggle_url(bc_url, {'Referer': url}),
1985                 'ie_key': 'BrightcoveLegacy'
1986             } for bc_url in bc_urls]
1987
1988             return {
1989                 '_type': 'playlist',
1990                 'title': video_title,
1991                 'id': video_id,
1992                 'entries': entries,
1993             }
1994
1995         # Look for Brightcove New Studio embeds
1996         bc_urls = BrightcoveNewIE._extract_urls(self, webpage)
1997         if bc_urls:
1998             return self.playlist_from_matches(bc_urls, video_id, video_title, ie='BrightcoveNew')
1999
2000         # Look for ThePlatform embeds
2001         tp_urls = ThePlatformIE._extract_urls(webpage)
2002         if tp_urls:
2003             return self.playlist_from_matches(tp_urls, video_id, video_title, ie='ThePlatform')
2004
2005         # Look for Vessel embeds
2006         vessel_urls = VesselIE._extract_urls(webpage)
2007         if vessel_urls:
2008             return self.playlist_from_matches(vessel_urls, video_id, video_title, ie=VesselIE.ie_key())
2009
2010         # Look for embedded rtl.nl player
2011         matches = re.findall(
2012             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
2013             webpage)
2014         if matches:
2015             return self.playlist_from_matches(matches, video_id, video_title, ie='RtlNl')
2016
2017         vimeo_urls = VimeoIE._extract_urls(url, webpage)
2018         if vimeo_urls:
2019             return self.playlist_from_matches(vimeo_urls, video_id, video_title, ie=VimeoIE.ie_key())
2020
2021         vid_me_embed_url = self._search_regex(
2022             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
2023             webpage, 'vid.me embed', default=None)
2024         if vid_me_embed_url is not None:
2025             return self.url_result(vid_me_embed_url, 'Vidme')
2026
2027         # Look for embedded YouTube player
2028         matches = re.findall(r'''(?x)
2029             (?:
2030                 <iframe[^>]+?src=|
2031                 data-video-url=|
2032                 <embed[^>]+?src=|
2033                 embedSWF\(?:\s*|
2034                 <object[^>]+data=|
2035                 new\s+SWFObject\(
2036             )
2037             (["\'])
2038                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2039                 (?:embed|v|p)/.+?)
2040             \1''', webpage)
2041         if matches:
2042             return self.playlist_from_matches(
2043                 matches, video_id, video_title, lambda m: unescapeHTML(m[1]))
2044
2045         # Look for lazyYT YouTube embed
2046         matches = re.findall(
2047             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
2048         if matches:
2049             return self.playlist_from_matches(matches, video_id, video_title, lambda m: unescapeHTML(m))
2050
2051         # Look for Wordpress "YouTube Video Importer" plugin
2052         matches = re.findall(r'''(?x)<div[^>]+
2053             class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2054             data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2055         if matches:
2056             return self.playlist_from_matches(matches, video_id, video_title, lambda m: m[-1])
2057
2058         matches = DailymotionIE._extract_urls(webpage)
2059         if matches:
2060             return self.playlist_from_matches(matches, video_id, video_title)
2061
2062         # Look for embedded Dailymotion playlist player (#3822)
2063         m = re.search(
2064             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
2065         if m:
2066             playlists = re.findall(
2067                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
2068             if playlists:
2069                 return self.playlist_from_matches(
2070                     playlists, video_id, video_title, lambda p: '//dailymotion.com/playlist/%s' % p)
2071
2072         # Look for embedded Wistia player
2073         match = re.search(
2074             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
2075         if match:
2076             embed_url = self._proto_relative_url(
2077                 unescapeHTML(match.group('url')))
2078             return {
2079                 '_type': 'url_transparent',
2080                 'url': embed_url,
2081                 'ie_key': 'Wistia',
2082                 'uploader': video_uploader,
2083             }
2084
2085         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
2086         if match:
2087             return {
2088                 '_type': 'url_transparent',
2089                 'url': 'wistia:%s' % match.group('id'),
2090                 'ie_key': 'Wistia',
2091                 'uploader': video_uploader,
2092             }
2093
2094         match = re.search(
2095             r'''(?sx)
2096                 <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*?
2097                 <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]+)\b.*?\2
2098             ''', webpage)
2099         if match:
2100             return self.url_result(self._proto_relative_url(
2101                 'wistia:%s' % match.group('id')), 'Wistia')
2102
2103         # Look for SVT player
2104         svt_url = SVTIE._extract_url(webpage)
2105         if svt_url:
2106             return self.url_result(svt_url, 'SVT')
2107
2108         # Look for embedded condenast player
2109         matches = re.findall(
2110             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
2111             webpage)
2112         if matches:
2113             return {
2114                 '_type': 'playlist',
2115                 'entries': [{
2116                     '_type': 'url',
2117                     'ie_key': 'CondeNast',
2118                     'url': ma,
2119                 } for ma in matches],
2120                 'title': video_title,
2121                 'id': video_id,
2122             }
2123
2124         # Look for Bandcamp pages with custom domain
2125         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
2126         if mobj is not None:
2127             burl = unescapeHTML(mobj.group(1))
2128             # Don't set the extractor because it can be a track url or an album
2129             return self.url_result(burl)
2130
2131         # Look for embedded Vevo player
2132         mobj = re.search(
2133             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
2134         if mobj is not None:
2135             return self.url_result(mobj.group('url'))
2136
2137         # Look for embedded Viddler player
2138         mobj = re.search(
2139             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
2140             webpage)
2141         if mobj is not None:
2142             return self.url_result(mobj.group('url'))
2143
2144         # Look for NYTimes player
2145         mobj = re.search(
2146             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
2147             webpage)
2148         if mobj is not None:
2149             return self.url_result(mobj.group('url'))
2150
2151         # Look for Libsyn player
2152         mobj = re.search(
2153             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
2154         if mobj is not None:
2155             return self.url_result(mobj.group('url'))
2156
2157         # Look for Ooyala videos
2158         mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
2159                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
2160                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
2161                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
2162         if mobj is not None:
2163             embed_token = self._search_regex(
2164                 r'embedToken[\'"]?\s*:\s*[\'"]([^\'"]+)',
2165                 webpage, 'ooyala embed token', default=None)
2166             return OoyalaIE._build_url_result(smuggle_url(
2167                 mobj.group('ec'), {
2168                     'domain': url,
2169                     'embed_token': embed_token,
2170                 }))
2171
2172         # Look for multiple Ooyala embeds on SBN network websites
2173         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
2174         if mobj is not None:
2175             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
2176             if embeds:
2177                 return self.playlist_from_matches(
2178                     embeds, video_id, video_title,
2179                     getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
2180
2181         # Look for Aparat videos
2182         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
2183         if mobj is not None:
2184             return self.url_result(mobj.group(1), 'Aparat')
2185
2186         # Look for MPORA videos
2187         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
2188         if mobj is not None:
2189             return self.url_result(mobj.group(1), 'Mpora')
2190
2191         # Look for embedded NovaMov-based player
2192         mobj = re.search(
2193             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
2194                     (?P<url>http://(?:(?:embed|www)\.)?
2195                         (?:novamov\.com|
2196                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
2197                            videoweed\.(?:es|com)|
2198                            movshare\.(?:net|sx|ag)|
2199                            divxstage\.(?:eu|net|ch|co|at|ag))
2200                         /embed\.php.+?)\1''', webpage)
2201         if mobj is not None:
2202             return self.url_result(mobj.group('url'))
2203
2204         # Look for embedded Facebook player
2205         facebook_url = FacebookIE._extract_url(webpage)
2206         if facebook_url is not None:
2207             return self.url_result(facebook_url, 'Facebook')
2208
2209         # Look for embedded VK player
2210         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
2211         if mobj is not None:
2212             return self.url_result(mobj.group('url'), 'VK')
2213
2214         # Look for embedded Odnoklassniki player
2215         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1', webpage)
2216         if mobj is not None:
2217             return self.url_result(mobj.group('url'), 'Odnoklassniki')
2218
2219         # Look for embedded ivi player
2220         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
2221         if mobj is not None:
2222             return self.url_result(mobj.group('url'), 'Ivi')
2223
2224         # Look for embedded Huffington Post player
2225         mobj = re.search(
2226             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
2227         if mobj is not None:
2228             return self.url_result(mobj.group('url'), 'HuffPost')
2229
2230         # Look for embed.ly
2231         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
2232         if mobj is not None:
2233             return self.url_result(mobj.group('url'))
2234         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
2235         if mobj is not None:
2236             return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
2237
2238         # Look for funnyordie embed
2239         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
2240         if matches:
2241             return self.playlist_from_matches(
2242                 matches, video_id, video_title, getter=unescapeHTML, ie='FunnyOrDie')
2243
2244         # Look for BBC iPlayer embed
2245         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
2246         if matches:
2247             return self.playlist_from_matches(matches, video_id, video_title, ie='BBCCoUk')
2248
2249         # Look for embedded RUTV player
2250         rutv_url = RUTVIE._extract_url(webpage)
2251         if rutv_url:
2252             return self.url_result(rutv_url, 'RUTV')
2253
2254         # Look for embedded TVC player
2255         tvc_url = TVCIE._extract_url(webpage)
2256         if tvc_url:
2257             return self.url_result(tvc_url, 'TVC')
2258
2259         # Look for embedded SportBox player
2260         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
2261         if sportbox_urls:
2262             return self.playlist_from_matches(sportbox_urls, video_id, video_title, ie='SportBoxEmbed')
2263
2264         # Look for embedded XHamster player
2265         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
2266         if xhamster_urls:
2267             return self.playlist_from_matches(xhamster_urls, video_id, video_title, ie='XHamsterEmbed')
2268
2269         # Look for embedded TNAFlixNetwork player
2270         tnaflix_urls = TNAFlixNetworkEmbedIE._extract_urls(webpage)
2271         if tnaflix_urls:
2272             return self.playlist_from_matches(tnaflix_urls, video_id, video_title, ie=TNAFlixNetworkEmbedIE.ie_key())
2273
2274         # Look for embedded PornHub player
2275         pornhub_urls = PornHubIE._extract_urls(webpage)
2276         if pornhub_urls:
2277             return self.playlist_from_matches(pornhub_urls, video_id, video_title, ie=PornHubIE.ie_key())
2278
2279         # Look for embedded DrTuber player
2280         drtuber_urls = DrTuberIE._extract_urls(webpage)
2281         if drtuber_urls:
2282             return self.playlist_from_matches(drtuber_urls, video_id, video_title, ie=DrTuberIE.ie_key())
2283
2284         # Look for embedded RedTube player
2285         redtube_urls = RedTubeIE._extract_urls(webpage)
2286         if redtube_urls:
2287             return self.playlist_from_matches(redtube_urls, video_id, video_title, ie=RedTubeIE.ie_key())
2288
2289         # Look for embedded Tvigle player
2290         mobj = re.search(
2291             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
2292         if mobj is not None:
2293             return self.url_result(mobj.group('url'), 'Tvigle')
2294
2295         # Look for embedded TED player
2296         mobj = re.search(
2297             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
2298         if mobj is not None:
2299             return self.url_result(mobj.group('url'), 'TED')
2300
2301         # Look for embedded Ustream videos
2302         ustream_url = UstreamIE._extract_url(webpage)
2303         if ustream_url:
2304             return self.url_result(ustream_url, UstreamIE.ie_key())
2305
2306         # Look for embedded arte.tv player
2307         mobj = re.search(
2308             r'<(?:script|iframe) [^>]*?src="(?P<url>http://www\.arte\.tv/(?:playerv2/embed|arte_vp/index)[^"]+)"',
2309             webpage)
2310         if mobj is not None:
2311             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
2312
2313         # Look for embedded francetv player
2314         mobj = re.search(
2315             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
2316             webpage)
2317         if mobj is not None:
2318             return self.url_result(mobj.group('url'))
2319
2320         # Look for embedded smotri.com player
2321         smotri_url = SmotriIE._extract_url(webpage)
2322         if smotri_url:
2323             return self.url_result(smotri_url, 'Smotri')
2324
2325         # Look for embedded Myvi.ru player
2326         myvi_url = MyviIE._extract_url(webpage)
2327         if myvi_url:
2328             return self.url_result(myvi_url)
2329
2330         # Look for embedded soundcloud player
2331         soundcloud_urls = SoundcloudIE._extract_urls(webpage)
2332         if soundcloud_urls:
2333             return self.playlist_from_matches(soundcloud_urls, video_id, video_title, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
2334
2335         # Look for tunein player
2336         tunein_urls = TuneInBaseIE._extract_urls(webpage)
2337         if tunein_urls:
2338             return self.playlist_from_matches(tunein_urls, video_id, video_title)
2339
2340         # Look for embedded mtvservices player
2341         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
2342         if mtvservices_url:
2343             return self.url_result(mtvservices_url, ie='MTVServicesEmbedded')
2344
2345         # Look for embedded yahoo player
2346         mobj = re.search(
2347             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
2348             webpage)
2349         if mobj is not None:
2350             return self.url_result(mobj.group('url'), 'Yahoo')
2351
2352         # Look for embedded sbs.com.au player
2353         mobj = re.search(
2354             r'''(?x)
2355             (?:
2356                 <meta\s+property="og:video"\s+content=|
2357                 <iframe[^>]+?src=
2358             )
2359             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
2360             webpage)
2361         if mobj is not None:
2362             return self.url_result(mobj.group('url'), 'SBS')
2363
2364         # Look for embedded Cinchcast player
2365         mobj = re.search(
2366             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
2367             webpage)
2368         if mobj is not None:
2369             return self.url_result(mobj.group('url'), 'Cinchcast')
2370
2371         mobj = re.search(
2372             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
2373             webpage)
2374         if not mobj:
2375             mobj = re.search(
2376                 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
2377                 webpage)
2378         if mobj is not None:
2379             return self.url_result(mobj.group('url'), 'MLB')
2380
2381         mobj = re.search(
2382             r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
2383             webpage)
2384         if mobj is not None:
2385             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
2386
2387         mobj = re.search(
2388             r'<iframe[^>]+src="(?P<url>https?://(?:new\.)?livestream\.com/[^"]+/player[^"]+)"',
2389             webpage)
2390         if mobj is not None:
2391             return self.url_result(mobj.group('url'), 'Livestream')
2392
2393         # Look for Zapiks embed
2394         mobj = re.search(
2395             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
2396         if mobj is not None:
2397             return self.url_result(mobj.group('url'), 'Zapiks')
2398
2399         # Look for Kaltura embeds
2400         kaltura_url = KalturaIE._extract_url(webpage)
2401         if kaltura_url:
2402             return self.url_result(smuggle_url(kaltura_url, {'source_url': url}), KalturaIE.ie_key())
2403
2404         # Look for Eagle.Platform embeds
2405         eagleplatform_url = EaglePlatformIE._extract_url(webpage)
2406         if eagleplatform_url:
2407             return self.url_result(eagleplatform_url, EaglePlatformIE.ie_key())
2408
2409         # Look for ClipYou (uses Eagle.Platform) embeds
2410         mobj = re.search(
2411             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
2412         if mobj is not None:
2413             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
2414
2415         # Look for Pladform embeds
2416         pladform_url = PladformIE._extract_url(webpage)
2417         if pladform_url:
2418             return self.url_result(pladform_url)
2419
2420         # Look for Videomore embeds
2421         videomore_url = VideomoreIE._extract_url(webpage)
2422         if videomore_url:
2423             return self.url_result(videomore_url)
2424
2425         # Look for Webcaster embeds
2426         webcaster_url = WebcasterFeedIE._extract_url(self, webpage)
2427         if webcaster_url:
2428             return self.url_result(webcaster_url, ie=WebcasterFeedIE.ie_key())
2429
2430         # Look for Playwire embeds
2431         mobj = re.search(
2432             r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
2433         if mobj is not None:
2434             return self.url_result(mobj.group('url'))
2435
2436         # Look for 5min embeds
2437         mobj = re.search(
2438             r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
2439         if mobj is not None:
2440             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
2441
2442         # Look for Crooks and Liars embeds
2443         mobj = re.search(
2444             r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
2445         if mobj is not None:
2446             return self.url_result(mobj.group('url'))
2447
2448         # Look for NBC Sports VPlayer embeds
2449         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
2450         if nbc_sports_url:
2451             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
2452
2453         # Look for NBC News embeds
2454         nbc_news_embed_url = re.search(
2455             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//www\.nbcnews\.com/widget/video-embed/[^"\']+)\1', webpage)
2456         if nbc_news_embed_url:
2457             return self.url_result(nbc_news_embed_url.group('url'), 'NBCNews')
2458
2459         # Look for Google Drive embeds
2460         google_drive_url = GoogleDriveIE._extract_url(webpage)
2461         if google_drive_url:
2462             return self.url_result(google_drive_url, 'GoogleDrive')
2463
2464         # Look for UDN embeds
2465         mobj = re.search(
2466             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage)
2467         if mobj is not None:
2468             return self.url_result(
2469                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
2470
2471         # Look for Senate ISVP iframe
2472         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
2473         if senate_isvp_url:
2474             return self.url_result(senate_isvp_url, 'SenateISVP')
2475
2476         # Look for Dailymotion Cloud videos
2477         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
2478         if dmcloud_url:
2479             return self.url_result(dmcloud_url, 'DailymotionCloud')
2480
2481         # Look for OnionStudios embeds
2482         onionstudios_url = OnionStudiosIE._extract_url(webpage)
2483         if onionstudios_url:
2484             return self.url_result(onionstudios_url)
2485
2486         # Look for ViewLift embeds
2487         viewlift_url = ViewLiftEmbedIE._extract_url(webpage)
2488         if viewlift_url:
2489             return self.url_result(viewlift_url)
2490
2491         # Look for JWPlatform embeds
2492         jwplatform_url = JWPlatformIE._extract_url(webpage)
2493         if jwplatform_url:
2494             return self.url_result(jwplatform_url, 'JWPlatform')
2495
2496         # Look for Digiteka embeds
2497         digiteka_url = DigitekaIE._extract_url(webpage)
2498         if digiteka_url:
2499             return self.url_result(self._proto_relative_url(digiteka_url), DigitekaIE.ie_key())
2500
2501         # Look for Arkena embeds
2502         arkena_url = ArkenaIE._extract_url(webpage)
2503         if arkena_url:
2504             return self.url_result(arkena_url, ArkenaIE.ie_key())
2505
2506         # Look for Piksel embeds
2507         piksel_url = PikselIE._extract_url(webpage)
2508         if piksel_url:
2509             return self.url_result(piksel_url, PikselIE.ie_key())
2510
2511         # Look for Limelight embeds
2512         limelight_urls = LimelightBaseIE._extract_urls(webpage, url)
2513         if limelight_urls:
2514             return self.playlist_result(
2515                 limelight_urls, video_id, video_title, video_description)
2516
2517         mobj = re.search(r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})', webpage)
2518         if mobj:
2519             lm = {
2520                 'Media': 'media',
2521                 'Channel': 'channel',
2522                 'ChannelList': 'channel_list',
2523             }
2524             return self.url_result(smuggle_url('limelight:%s:%s' % (
2525                 lm[mobj.group(1)], mobj.group(2)), {'source_url': url}),
2526                 'Limelight%s' % mobj.group(1), mobj.group(2))
2527
2528         mobj = re.search(
2529             r'''(?sx)
2530                 <object[^>]+class=(["\'])LimelightEmbeddedPlayerFlash\1[^>]*>.*?
2531                     <param[^>]+
2532                         name=(["\'])flashVars\2[^>]+
2533                         value=(["\'])(?:(?!\3).)*mediaId=(?P<id>[a-z0-9]{32})
2534             ''', webpage)
2535         if mobj:
2536             return self.url_result(smuggle_url(
2537                 'limelight:media:%s' % mobj.group('id'),
2538                 {'source_url': url}), 'LimelightMedia', mobj.group('id'))
2539
2540         # Look for AdobeTVVideo embeds
2541         mobj = re.search(
2542             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
2543             webpage)
2544         if mobj is not None:
2545             return self.url_result(
2546                 self._proto_relative_url(unescapeHTML(mobj.group(1))),
2547                 'AdobeTVVideo')
2548
2549         # Look for Vine embeds
2550         mobj = re.search(
2551             r'<iframe[^>]+src=[\'"]((?:https?:)?//(?:www\.)?vine\.co/v/[^/]+/embed/(?:simple|postcard))',
2552             webpage)
2553         if mobj is not None:
2554             return self.url_result(
2555                 self._proto_relative_url(unescapeHTML(mobj.group(1))), 'Vine')
2556
2557         # Look for VODPlatform embeds
2558         mobj = re.search(
2559             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vod-platform\.net/[eE]mbed/.+?)\1',
2560             webpage)
2561         if mobj is not None:
2562             return self.url_result(
2563                 self._proto_relative_url(unescapeHTML(mobj.group('url'))), 'VODPlatform')
2564
2565         # Look for Mangomolo embeds
2566         mobj = re.search(
2567             r'''(?x)<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?admin\.mangomolo\.com/analytics/index\.php/customers/embed/
2568                 (?:
2569                     video\?.*?\bid=(?P<video_id>\d+)|
2570                     index\?.*?\bchannelid=(?P<channel_id>(?:[A-Za-z0-9+/=]|%2B|%2F|%3D)+)
2571                 ).+?)\1''', webpage)
2572         if mobj is not None:
2573             info = {
2574                 '_type': 'url_transparent',
2575                 'url': self._proto_relative_url(unescapeHTML(mobj.group('url'))),
2576                 'title': video_title,
2577                 'description': video_description,
2578                 'thumbnail': video_thumbnail,
2579                 'uploader': video_uploader,
2580             }
2581             video_id = mobj.group('video_id')
2582             if video_id:
2583                 info.update({
2584                     'ie_key': 'MangomoloVideo',
2585                     'id': video_id,
2586                 })
2587             else:
2588                 info.update({
2589                     'ie_key': 'MangomoloLive',
2590                     'id': mobj.group('channel_id'),
2591                 })
2592             return info
2593
2594         # Look for Instagram embeds
2595         instagram_embed_url = InstagramIE._extract_embed_url(webpage)
2596         if instagram_embed_url is not None:
2597             return self.url_result(
2598                 self._proto_relative_url(instagram_embed_url), InstagramIE.ie_key())
2599
2600         # Look for LiveLeak embeds
2601         liveleak_url = LiveLeakIE._extract_url(webpage)
2602         if liveleak_url:
2603             return self.url_result(liveleak_url, 'LiveLeak')
2604
2605         # Look for 3Q SDN embeds
2606         threeqsdn_url = ThreeQSDNIE._extract_url(webpage)
2607         if threeqsdn_url:
2608             return {
2609                 '_type': 'url_transparent',
2610                 'ie_key': ThreeQSDNIE.ie_key(),
2611                 'url': self._proto_relative_url(threeqsdn_url),
2612                 'title': video_title,
2613                 'description': video_description,
2614                 'thumbnail': video_thumbnail,
2615                 'uploader': video_uploader,
2616             }
2617
2618         # Look for VBOX7 embeds
2619         vbox7_url = Vbox7IE._extract_url(webpage)
2620         if vbox7_url:
2621             return self.url_result(vbox7_url, Vbox7IE.ie_key())
2622
2623         # Look for DBTV embeds
2624         dbtv_urls = DBTVIE._extract_urls(webpage)
2625         if dbtv_urls:
2626             return self.playlist_from_matches(dbtv_urls, video_id, video_title, ie=DBTVIE.ie_key())
2627
2628         # Look for Videa embeds
2629         videa_urls = VideaIE._extract_urls(webpage)
2630         if videa_urls:
2631             return self.playlist_from_matches(videa_urls, video_id, video_title, ie=VideaIE.ie_key())
2632
2633         # Look for 20 minuten embeds
2634         twentymin_urls = TwentyMinutenIE._extract_urls(webpage)
2635         if twentymin_urls:
2636             return self.playlist_from_matches(
2637                 twentymin_urls, video_id, video_title, ie=TwentyMinutenIE.ie_key())
2638
2639         # Look for Openload embeds
2640         openload_urls = OpenloadIE._extract_urls(webpage)
2641         if openload_urls:
2642             return self.playlist_from_matches(
2643                 openload_urls, video_id, video_title, ie=OpenloadIE.ie_key())
2644
2645         # Look for VideoPress embeds
2646         videopress_urls = VideoPressIE._extract_urls(webpage)
2647         if videopress_urls:
2648             return self.playlist_from_matches(
2649                 videopress_urls, video_id, video_title, ie=VideoPressIE.ie_key())
2650
2651         # Look for Rutube embeds
2652         rutube_urls = RutubeIE._extract_urls(webpage)
2653         if rutube_urls:
2654             return self.playlist_from_matches(
2655                 rutube_urls, ie=RutubeIE.ie_key())
2656
2657         # Looking for http://schema.org/VideoObject
2658         json_ld = self._search_json_ld(
2659             webpage, video_id, default={}, expected_type='VideoObject')
2660         if json_ld.get('url'):
2661             info_dict.update({
2662                 'title': video_title or info_dict['title'],
2663                 'description': video_description,
2664                 'thumbnail': video_thumbnail,
2665                 'age_limit': age_limit
2666             })
2667             info_dict.update(json_ld)
2668             return info_dict
2669
2670         # Look for HTML5 media
2671         entries = self._parse_html5_media_entries(url, webpage, video_id, m3u8_id='hls')
2672         if entries:
2673             for entry in entries:
2674                 entry.update({
2675                     'id': video_id,
2676                     'title': video_title,
2677                 })
2678                 self._sort_formats(entry['formats'])
2679             return self.playlist_result(entries)
2680
2681         jwplayer_data = self._find_jwplayer_data(
2682             webpage, video_id, transform_source=js_to_json)
2683         if jwplayer_data:
2684             info = self._parse_jwplayer_data(
2685                 jwplayer_data, video_id, require_title=False, base_url=url)
2686             if not info.get('title'):
2687                 info['title'] = video_title
2688             return info
2689
2690         def check_video(vurl):
2691             if YoutubeIE.suitable(vurl):
2692                 return True
2693             if RtmpIE.suitable(vurl):
2694                 return True
2695             vpath = compat_urlparse.urlparse(vurl).path
2696             vext = determine_ext(vpath)
2697             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml', 'js', 'xml')
2698
2699         def filter_video(urls):
2700             return list(filter(check_video, urls))
2701
2702         # Start with something easy: JW Player in SWFObject
2703         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
2704         if not found:
2705             # Look for gorilla-vid style embedding
2706             found = filter_video(re.findall(r'''(?sx)
2707                 (?:
2708                     jw_plugins|
2709                     JWPlayerOptions|
2710                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
2711                 )
2712                 .*?
2713                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
2714         if not found:
2715             # Broaden the search a little bit
2716             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
2717         if not found:
2718             # Broaden the findall a little bit: JWPlayer JS loader
2719             found = filter_video(re.findall(
2720                 r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
2721         if not found:
2722             # Flow player
2723             found = filter_video(re.findall(r'''(?xs)
2724                 flowplayer\("[^"]+",\s*
2725                     \{[^}]+?\}\s*,
2726                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
2727                         ["']?url["']?\s*:\s*["']([^"']+)["']
2728             ''', webpage))
2729         if not found:
2730             # Cinerama player
2731             found = re.findall(
2732                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
2733         if not found:
2734             # Try to find twitter cards info
2735             # twitter:player:stream should be checked before twitter:player since
2736             # it is expected to contain a raw stream (see
2737             # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser)
2738             found = filter_video(re.findall(
2739                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
2740         if not found:
2741             # We look for Open Graph info:
2742             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
2743             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
2744             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
2745             if m_video_type is not None:
2746                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
2747         if not found:
2748             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
2749             found = re.search(
2750                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
2751                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
2752                 webpage)
2753             if not found:
2754                 # Look also in Refresh HTTP header
2755                 refresh_header = head_response.headers.get('Refresh')
2756                 if refresh_header:
2757                     # In python 2 response HTTP headers are bytestrings
2758                     if sys.version_info < (3, 0) and isinstance(refresh_header, str):
2759                         refresh_header = refresh_header.decode('iso-8859-1')
2760                     found = re.search(REDIRECT_REGEX, refresh_header)
2761             if found:
2762                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
2763                 if new_url != url:
2764                     self.report_following_redirect(new_url)
2765                     return {
2766                         '_type': 'url',
2767                         'url': new_url,
2768                     }
2769                 else:
2770                     found = None
2771
2772         if not found:
2773             # twitter:player is a https URL to iframe player that may or may not
2774             # be supported by youtube-dl thus this is checked the very last (see
2775             # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser)
2776             embed_url = self._html_search_meta('twitter:player', webpage, default=None)
2777             if embed_url:
2778                 return self.url_result(embed_url)
2779
2780         if not found:
2781             raise UnsupportedError(url)
2782
2783         entries = []
2784         for video_url in orderedSet(found):
2785             video_url = unescapeHTML(video_url)
2786             video_url = video_url.replace('\\/', '/')
2787             video_url = compat_urlparse.urljoin(url, video_url)
2788             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
2789
2790             # Sometimes, jwplayer extraction will result in a YouTube URL
2791             if YoutubeIE.suitable(video_url):
2792                 entries.append(self.url_result(video_url, 'Youtube'))
2793                 continue
2794
2795             # here's a fun little line of code for you:
2796             video_id = os.path.splitext(video_id)[0]
2797
2798             entry_info_dict = {
2799                 'id': video_id,
2800                 'uploader': video_uploader,
2801                 'title': video_title,
2802                 'age_limit': age_limit,
2803             }
2804
2805             if RtmpIE.suitable(video_url):
2806                 entry_info_dict.update({
2807                     '_type': 'url_transparent',
2808                     'ie_key': RtmpIE.ie_key(),
2809                     'url': video_url,
2810                 })
2811                 entries.append(entry_info_dict)
2812                 continue
2813
2814             ext = determine_ext(video_url)
2815             if ext == 'smil':
2816                 entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
2817             elif ext == 'xspf':
2818                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
2819             elif ext == 'm3u8':
2820                 entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
2821             elif ext == 'mpd':
2822                 entry_info_dict['formats'] = self._extract_mpd_formats(video_url, video_id)
2823             elif ext == 'f4m':
2824                 entry_info_dict['formats'] = self._extract_f4m_formats(video_url, video_id)
2825             elif re.search(r'(?i)\.(?:ism|smil)/manifest', video_url) and video_url != url:
2826                 # Just matching .ism/manifest is not enough to be reliably sure
2827                 # whether it's actually an ISM manifest or some other streaming
2828                 # manifest since there are various streaming URL formats
2829                 # possible (see [1]) as well as some other shenanigans like
2830                 # .smil/manifest URLs that actually serve an ISM (see [2]) and
2831                 # so on.
2832                 # Thus the most reasonable way to solve this is to delegate
2833                 # to generic extractor in order to look into the contents of
2834                 # the manifest itself.
2835                 # 1. https://azure.microsoft.com/en-us/documentation/articles/media-services-deliver-content-overview/#streaming-url-formats
2836                 # 2. https://svs.itworkscdn.net/lbcivod/smil:itwfcdn/lbci/170976.smil/Manifest
2837                 entry_info_dict = self.url_result(
2838                     smuggle_url(video_url, {'to_generic': True}),
2839                     GenericIE.ie_key())
2840             else:
2841                 entry_info_dict['url'] = video_url
2842
2843             if entry_info_dict.get('formats'):
2844                 self._sort_formats(entry_info_dict['formats'])
2845
2846             entries.append(entry_info_dict)
2847
2848         if len(entries) == 1:
2849             return entries[0]
2850         else:
2851             for num, e in enumerate(entries, start=1):
2852                 # 'url' results don't have a title
2853                 if e.get('title') is not None:
2854                     e['title'] = '%s (%d)' % (e['title'], num)
2855             return {
2856                 '_type': 'playlist',
2857                 'entries': entries,
2858             }