_ 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_str,
  14     compat_urllib_parse_unquote,
  15     compat_urlparse,
  16     compat_xml_parse_error,
  17 )
  18 from ..utils import (
  19     determine_ext,
  20     ExtractorError,
  21     float_or_none,
  22     HEADRequest,
  23     is_html,
  24     js_to_json,
  25     orderedSet,
  26     sanitized_Request,
  27     smuggle_url,
  28     unescapeHTML,
  29     unified_strdate,
  30     unsmuggle_url,
  31     UnsupportedError,
  32     xpath_text,
  33 )
  34 from .commonprotocols import RtmpIE
  35 from .brightcove import (
  36     BrightcoveLegacyIE,
  37     BrightcoveNewIE,
  38 )
  39 from .nbc import NBCSportsVPlayerIE
  40 from .ooyala import OoyalaIE
  41 from .rutv import RUTVIE
  42 from .tvc import TVCIE
  43 from .sportbox import SportBoxEmbedIE
  44 from .smotri import SmotriIE
  45 from .myvi import MyviIE
  46 from .condenast import CondeNastIE
  47 from .udn import UDNEmbedIE
  48 from .senateisvp import SenateISVPIE
  49 from .svt import SVTIE
  50 from .pornhub import PornHubIE
  51 from .xhamster import XHamsterEmbedIE
  52 from .tnaflix import TNAFlixNetworkEmbedIE
  53 from .drtuber import DrTuberIE
  54 from .redtube import RedTubeIE
  55 from .vimeo import VimeoIE
  56 from .dailymotion import (
  57     DailymotionIE,
  58     DailymotionCloudIE,
  59 )
  60 from .onionstudios import OnionStudiosIE
  61 from .viewlift import ViewLiftEmbedIE
  62 from .mtv import MTVServicesEmbeddedIE
  63 from .pladform import PladformIE
  64 from .videomore import VideomoreIE
  65 from .webcaster import WebcasterFeedIE
  66 from .googledrive import GoogleDriveIE
  67 from .jwplatform import JWPlatformIE
  68 from .digiteka import DigitekaIE
  69 from .arkena import ArkenaIE
  70 from .instagram import InstagramIE
  71 from .liveleak import LiveLeakIE
  72 from .threeqsdn import ThreeQSDNIE
  73 from .theplatform import ThePlatformIE
  74 from .vessel import VesselIE
  75 from .kaltura import KalturaIE
  76 from .eagleplatform import EaglePlatformIE
  77 from .facebook import FacebookIE
  78 from .soundcloud import SoundcloudIE
  79 from .tunein import TuneInBaseIE
  80 from .vbox7 import Vbox7IE
  81 from .dbtv import DBTVIE
  82 from .piksel import PikselIE
  83 from .videa import VideaIE
  84 from .twentymin import TwentyMinutenIE
  85 from .ustream import UstreamIE
  86 from .openload import OpenloadIE
  87 from .videopress import VideoPressIE
  88 from .rutube import RutubeIE
  89 from .limelight import LimelightBaseIE
  90 from .anvato import AnvatoIE
  91 from .washingtonpost import WashingtonPostIE
  92 from .wistia import WistiaIE
  93 from .mediaset import MediasetIE
  94
  95
  96 class GenericIE(InfoExtractor):
  97     IE_DESC = 'Generic downloader that works on some sites'
  98     _VALID_URL = r'.*'
  99     IE_NAME = 'generic'
 100     _TESTS = [
 101         # Direct link to a video
 102         {
 103             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
 104             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
 105             'info_dict': {
 106                 'id': 'trailer',
 107                 'ext': 'mp4',
 108                 'title': 'trailer',
 109                 'upload_date': '20100513',
 110             }
 111         },
 112         # Direct link to media delivered compressed (until Accept-Encoding is *)
 113         {
 114             'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
 115             'md5': '128c42e68b13950268b648275386fc74',
 116             'info_dict': {
 117                 'id': 'FictionJunction-Parallel_Hearts',
 118                 'ext': 'flac',
 119                 'title': 'FictionJunction-Parallel_Hearts',
 120                 'upload_date': '20140522',
 121             },
 122             'expected_warnings': [
 123                 'URL could be a direct video link, returning it as such.'
 124             ],
 125             'skip': 'URL invalid',
 126         },
 127         # Direct download with broken HEAD
 128         {
 129             'url': 'http://ai-radio.org:8000/radio.opus',
 130             'info_dict': {
 131                 'id': 'radio',
 132                 'ext': 'opus',
 133                 'title': 'radio',
 134             },
 135             'params': {
 136                 'skip_download': True,  # infinite live stream
 137             },
 138             'expected_warnings': [
 139                 r'501.*Not Implemented',
 140                 r'400.*Bad Request',
 141             ],
 142         },
 143         # Direct link with incorrect MIME type
 144         {
 145             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 146             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 147             'info_dict': {
 148                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 149                 'id': '5_Lennart_Poettering_-_Systemd',
 150                 'ext': 'webm',
 151                 'title': '5_Lennart_Poettering_-_Systemd',
 152                 'upload_date': '20141120',
 153             },
 154             'expected_warnings': [
 155                 'URL could be a direct video link, returning it as such.'
 156             ]
 157         },
 158         # RSS feed
 159         {
 160             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 161             'info_dict': {
 162                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 163                 'title': 'Zero Punctuation',
 164                 'description': 're:.*groundbreaking video review series.*'
 165             },
 166             'playlist_mincount': 11,
 167         },
 168         # RSS feed with enclosure
 169         {
 170             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
 171             'info_dict': {
 172                 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 173                 'ext': 'm4v',
 174                 'upload_date': '20150228',
 175                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
 176             }
 177         },
 178         # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
 179         {
 180             'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
 181             'info_dict': {
 182                 'id': 'smil',
 183                 'ext': 'mp4',
 184                 'title': 'Automatics, robotics and biocybernetics',
 185                 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
 186                 'upload_date': '20130627',
 187                 'formats': 'mincount:16',
 188                 'subtitles': 'mincount:1',
 189             },
 190             'params': {
 191                 'force_generic_extractor': True,
 192                 'skip_download': True,
 193             },
 194         },
 195         # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html
 196         {
 197             'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil',
 198             'info_dict': {
 199                 'id': 'hds',
 200                 'ext': 'flv',
 201                 'title': 'hds',
 202                 'formats': 'mincount:1',
 203             },
 204             'params': {
 205                 'skip_download': True,
 206             },
 207         },
 208         # SMIL from https://www.restudy.dk/video/play/id/1637
 209         {
 210             'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml',
 211             'info_dict': {
 212                 'id': 'video_1637',
 213                 'ext': 'flv',
 214                 'title': 'video_1637',
 215                 'formats': 'mincount:3',
 216             },
 217             'params': {
 218                 'skip_download': True,
 219             },
 220         },
 221         # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm
 222         {
 223             'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil',
 224             'info_dict': {
 225                 'id': 'smil-service',
 226                 'ext': 'flv',
 227                 'title': 'smil-service',
 228                 'formats': 'mincount:1',
 229             },
 230             'params': {
 231                 'skip_download': True,
 232             },
 233         },
 234         # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370
 235         {
 236             'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil',
 237             'info_dict': {
 238                 'id': '4719370',
 239                 'ext': 'mp4',
 240                 'title': '571de1fd-47bc-48db-abf9-238872a58d1f',
 241                 'formats': 'mincount:3',
 242             },
 243             'params': {
 244                 'skip_download': True,
 245             },
 246         },
 247         # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html
 248         {
 249             'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf',
 250             'info_dict': {
 251                 'id': 'mZlp2ctYIUEB',
 252                 'ext': 'mp4',
 253                 'title': 'Tikibad ontruimd wegens brand',
 254                 'description': 'md5:05ca046ff47b931f9b04855015e163a4',
 255                 'thumbnail': r're:^https?://.*\.jpg$',
 256                 'duration': 33,
 257             },
 258             'params': {
 259                 'skip_download': True,
 260             },
 261         },
 262         # MPD from http://dash-mse-test.appspot.com/media.html
 263         {
 264             'url': 'http://yt-dash-mse-test.commondatastorage.googleapis.com/media/car-20120827-manifest.mpd',
 265             'md5': '4b57baab2e30d6eb3a6a09f0ba57ef53',
 266             'info_dict': {
 267                 'id': 'car-20120827-manifest',
 268                 'ext': 'mp4',
 269                 'title': 'car-20120827-manifest',
 270                 'formats': 'mincount:9',
 271                 'upload_date': '20130904',
 272             },
 273             'params': {
 274                 'format': 'bestvideo',
 275             },
 276         },
 277         # m3u8 served with Content-Type: audio/x-mpegURL; charset=utf-8
 278         {
 279             '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',
 280             'info_dict': {
 281                 'id': 'content',
 282                 'ext': 'mp4',
 283                 'title': 'content',
 284                 'formats': 'mincount:8',
 285             },
 286             'params': {
 287                 # m3u8 downloads
 288                 'skip_download': True,
 289             },
 290             'skip': 'video gone',
 291         },
 292         # m3u8 served with Content-Type: text/plain
 293         {
 294             'url': 'http://www.nacentapps.com/m3u8/index.m3u8',
 295             'info_dict': {
 296                 'id': 'index',
 297                 'ext': 'mp4',
 298                 'title': 'index',
 299                 'upload_date': '20140720',
 300                 'formats': 'mincount:11',
 301             },
 302             'params': {
 303                 # m3u8 downloads
 304                 'skip_download': True,
 305             },
 306             'skip': 'video gone',
 307         },
 308         # google redirect
 309         {
 310             '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',
 311             'info_dict': {
 312                 'id': 'cmQHVoWB5FY',
 313                 'ext': 'mp4',
 314                 'upload_date': '20130224',
 315                 'uploader_id': 'TheVerge',
 316                 'description': r're:^Chris Ziegler takes a look at the\.*',
 317                 'uploader': 'The Verge',
 318                 'title': 'First Firefox OS phones side-by-side',
 319             },
 320             'params': {
 321                 'skip_download': False,
 322             }
 323         },
 324         {
 325             # redirect in Refresh HTTP header
 326             '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',
 327             'info_dict': {
 328                 'id': 'pO8h3EaFRdo',
 329                 'ext': 'mp4',
 330                 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
 331                 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5',
 332                 'upload_date': '20150917',
 333                 'uploader_id': 'brtvofficial',
 334                 'uploader': 'Boiler Room',
 335             },
 336             'params': {
 337                 'skip_download': False,
 338             },
 339         },
 340         {
 341             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
 342             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
 343             'info_dict': {
 344                 'id': '13601338388002',
 345                 'ext': 'mp4',
 346                 'uploader': 'www.hodiho.fr',
 347                 'title': 'R\u00e9gis plante sa Jeep',
 348             }
 349         },
 350         # bandcamp page with custom domain
 351         {
 352             'add_ie': ['Bandcamp'],
 353             'url': 'http://bronyrock.com/track/the-pony-mash',
 354             'info_dict': {
 355                 'id': '3235767654',
 356                 'ext': 'mp3',
 357                 'title': 'The Pony Mash',
 358                 'uploader': 'M_Pallante',
 359             },
 360             'skip': 'There is a limit of 200 free downloads / month for the test song',
 361         },
 362         {
 363             # embedded brightcove video
 364             # it also tests brightcove videos that need to set the 'Referer'
 365             # in the http requests
 366             'add_ie': ['BrightcoveLegacy'],
 367             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
 368             'info_dict': {
 369                 'id': '2765128793001',
 370                 'ext': 'mp4',
 371                 'title': 'Le cours de bourse : l’analyse technique',
 372                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
 373                 'uploader': 'BFM BUSINESS',
 374             },
 375             'params': {
 376                 'skip_download': True,
 377             },
 378         },
 379         {
 380             # embedded with itemprop embedURL and video id spelled as `idVideo`
 381             'add_id': ['BrightcoveLegacy'],
 382             'url': 'http://bfmbusiness.bfmtv.com/mediaplayer/chroniques/olivier-delamarche/',
 383             'info_dict': {
 384                 'id': '5255628253001',
 385                 'ext': 'mp4',
 386                 'title': 'md5:37c519b1128915607601e75a87995fc0',
 387                 'description': 'md5:37f7f888b434bb8f8cc8dbd4f7a4cf26',
 388                 'uploader': 'BFM BUSINESS',
 389                 'uploader_id': '876450612001',
 390                 'timestamp': 1482255315,
 391                 'upload_date': '20161220',
 392             },
 393             'params': {
 394                 'skip_download': True,
 395             },
 396         },
 397         {
 398             # https://github.com/rg3/youtube-dl/issues/2253
 399             'url': 'http://bcove.me/i6nfkrc3',
 400             'md5': '0ba9446db037002366bab3b3eb30c88c',
 401             'info_dict': {
 402                 'id': '3101154703001',
 403                 'ext': 'mp4',
 404                 'title': 'Still no power',
 405                 'uploader': 'thestar.com',
 406                 '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.',
 407             },
 408             'add_ie': ['BrightcoveLegacy'],
 409             'skip': 'video gone',
 410         },
 411         {
 412             'url': 'http://www.championat.com/video/football/v/87/87499.html',
 413             'md5': 'fb973ecf6e4a78a67453647444222983',
 414             'info_dict': {
 415                 'id': '3414141473001',
 416                 'ext': 'mp4',
 417                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 418                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 419                 'uploader': 'Championat',
 420             },
 421         },
 422         {
 423             # https://github.com/rg3/youtube-dl/issues/3541
 424             'add_ie': ['BrightcoveLegacy'],
 425             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 426             'info_dict': {
 427                 'id': '3866516442001',
 428                 'ext': 'mp4',
 429                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 430                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 431                 'uploader': 'SBS Broadcasting',
 432             },
 433             'skip': 'Restricted to Netherlands',
 434             'params': {
 435                 'skip_download': True,  # m3u8 download
 436             },
 437         },
 438         {
 439             # Brightcove video in <iframe>
 440             'url': 'http://www.un.org/chinese/News/story.asp?NewsID=27724',
 441             'md5': '36d74ef5e37c8b4a2ce92880d208b968',
 442             'info_dict': {
 443                 'id': '5360463607001',
 444                 'ext': 'mp4',
 445                 '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',
 446                 '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。',
 447                 'uploader': 'United Nations',
 448                 'uploader_id': '1362235914001',
 449                 'timestamp': 1489593889,
 450                 'upload_date': '20170315',
 451             },
 452             'add_ie': ['BrightcoveLegacy'],
 453         },
 454         {
 455             # Brightcove with alternative playerID key
 456             'url': 'http://www.nature.com/nmeth/journal/v9/n7/fig_tab/nmeth.2062_SV1.html',
 457             'info_dict': {
 458                 'id': 'nmeth.2062_SV1',
 459                 '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',
 460             },
 461             'playlist': [{
 462                 'info_dict': {
 463                     'id': '2228375078001',
 464                     'ext': 'mp4',
 465                     'title': 'nmeth.2062-sv1',
 466                     'description': 'nmeth.2062-sv1',
 467                     'timestamp': 1363357591,
 468                     'upload_date': '20130315',
 469                     'uploader': 'Nature Publishing Group',
 470                     'uploader_id': '1964492299001',
 471                 },
 472             }],
 473         },
 474         {
 475             # Brightcove with UUID in videoPlayer
 476             'url': 'http://www8.hp.com/cn/zh/home.html',
 477             'info_dict': {
 478                 'id': '5255815316001',
 479                 'ext': 'mp4',
 480                 'title': 'Sprocket Video - China',
 481                 'description': 'Sprocket Video - China',
 482                 'uploader': 'HP-Video Gallery',
 483                 'timestamp': 1482263210,
 484                 'upload_date': '20161220',
 485                 'uploader_id': '1107601872001',
 486             },
 487             'params': {
 488                 'skip_download': True,  # m3u8 download
 489             },
 490             'skip': 'video rotates...weekly?',
 491         },
 492         {
 493             # Brightcove:new type [2].
 494             'url': 'http://www.delawaresportszone.com/video-st-thomas-more-earns-first-trip-to-basketball-semis',
 495             'md5': '2b35148fcf48da41c9fb4591650784f3',
 496             'info_dict': {
 497                 'id': '5348741021001',
 498                 'ext': 'mp4',
 499                 'upload_date': '20170306',
 500                 'uploader_id': '4191638492001',
 501                 'timestamp': 1488769918,
 502                 'title': 'VIDEO:  St. Thomas More earns first trip to basketball semis',
 503
 504             },
 505         },
 506         {
 507             # Alternative brightcove <video> attributes
 508             '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/',
 509             'info_dict': {
 510                 'id': '81095-guillaume-canet-evoque-les-rumeurs-d-infidelite-de-marion-cotillard-avec-brad-pitt-dans-vivement-dimanche',
 511                 'title': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche, Extraits : toutes les vidéos avec Télé-Loisirs",
 512             },
 513             'playlist': [{
 514                 'md5': '732d22ba3d33f2f3fc253c39f8f36523',
 515                 'info_dict': {
 516                     'id': '5311302538001',
 517                     'ext': 'mp4',
 518                     'title': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche",
 519                     'description': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche (France 2, 5 février 2017)",
 520                     'timestamp': 1486321708,
 521                     'upload_date': '20170205',
 522                     'uploader_id': '800000640001',
 523                 },
 524                 'only_matching': True,
 525             }],
 526         },
 527         {
 528             # Brightcove with UUID in videoPlayer
 529             'url': 'http://www8.hp.com/cn/zh/home.html',
 530             'info_dict': {
 531                 'id': '5255815316001',
 532                 'ext': 'mp4',
 533                 'title': 'Sprocket Video - China',
 534                 'description': 'Sprocket Video - China',
 535                 'uploader': 'HP-Video Gallery',
 536                 'timestamp': 1482263210,
 537                 'upload_date': '20161220',
 538                 'uploader_id': '1107601872001',
 539             },
 540             'params': {
 541                 'skip_download': True,  # m3u8 download
 542             },
 543         },
 544         # ooyala video
 545         {
 546             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 547             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 548             'info_dict': {
 549                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 550                 'ext': 'mp4',
 551                 'title': '2cc213299525360.mov',  # that's what we get
 552                 'duration': 238.231,
 553             },
 554             'add_ie': ['Ooyala'],
 555         },
 556         {
 557             # ooyala video embedded with http://player.ooyala.com/iframe.js
 558             'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/',
 559             'info_dict': {
 560                 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB',
 561                 'ext': 'mp4',
 562                 'title': '"Steve Jobs: Man in the Machine" trailer',
 563                 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
 564                 'duration': 135.427,
 565             },
 566             'params': {
 567                 'skip_download': True,
 568             },
 569             'skip': 'movie expired',
 570         },
 571         # embed.ly video
 572         {
 573             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 574             'info_dict': {
 575                 'id': '9ODmcdjQcHQ',
 576                 'ext': 'mp4',
 577                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 578                 'upload_date': '20140225',
 579                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 580                 'uploader': 'Tested',
 581                 'uploader_id': 'testedcom',
 582             },
 583             # No need to test YoutubeIE here
 584             'params': {
 585                 'skip_download': True,
 586             },
 587         },
 588         # funnyordie embed
 589         {
 590             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 591             'info_dict': {
 592                 'id': '18e820ec3f',
 593                 'ext': 'mp4',
 594                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 595                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 596             },
 597             # HEAD requests lead to endless 301, while GET is OK
 598             'expected_warnings': ['301'],
 599         },
 600         # RUTV embed
 601         {
 602             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 603             'info_dict': {
 604                 'id': '776940',
 605                 'ext': 'mp4',
 606                 'title': 'Охотское море стало целиком российским',
 607                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 608             },
 609             'params': {
 610                 # m3u8 download
 611                 'skip_download': True,
 612             },
 613         },
 614         # TVC embed
 615         {
 616             '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/',
 617             'info_dict': {
 618                 'id': '55304',
 619                 'ext': 'mp4',
 620                 'title': 'Дошкольное воспитание',
 621             },
 622         },
 623         # SportBox embed
 624         {
 625             'url': 'http://www.vestifinance.ru/articles/25753',
 626             'info_dict': {
 627                 'id': '25753',
 628                 'title': 'Прямые трансляции с Форума-выставки "Госзаказ-2013"',
 629             },
 630             'playlist': [{
 631                 'info_dict': {
 632                     'id': '370908',
 633                     'title': 'Госзаказ. День 3',
 634                     'ext': 'mp4',
 635                 }
 636             }, {
 637                 'info_dict': {
 638                     'id': '370905',
 639                     'title': 'Госзаказ. День 2',
 640                     'ext': 'mp4',
 641                 }
 642             }, {
 643                 'info_dict': {
 644                     'id': '370902',
 645                     'title': 'Госзаказ. День 1',
 646                     'ext': 'mp4',
 647                 }
 648             }],
 649             'params': {
 650                 # m3u8 download
 651                 'skip_download': True,
 652             },
 653         },
 654         # Myvi.ru embed
 655         {
 656             'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
 657             'info_dict': {
 658                 'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
 659                 'ext': 'mp4',
 660                 'title': 'Ужастики, русский трейлер (2015)',
 661                 'thumbnail': r're:^https?://.*\.jpg$',
 662                 'duration': 153,
 663             }
 664         },
 665         # XHamster embed
 666         {
 667             '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',
 668             'info_dict': {
 669                 'id': 'showthread',
 670                 'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
 671             },
 672             'playlist_mincount': 7,
 673             # This forum does not allow <iframe> syntaxes anymore
 674             # Now HTML tags are displayed as-is
 675             'skip': 'No videos on this page',
 676         },
 677         # Embedded TED video
 678         {
 679             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 680             'md5': '65fdff94098e4a607385a60c5177c638',
 681             'info_dict': {
 682                 'id': '1969',
 683                 'ext': 'mp4',
 684                 'title': 'Hidden miracles of the natural world',
 685                 'uploader': 'Louie Schwartzberg',
 686                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 687             }
 688         },
 689         # nowvideo embed hidden behind percent encoding
 690         {
 691             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 692             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 693             'info_dict': {
 694                 'id': '06e53103ca9aa',
 695                 'ext': 'flv',
 696                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 697                 'description': 'No description',
 698             },
 699         },
 700         # arte embed
 701         {
 702             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 703             'md5': '7653032cbb25bf6c80d80f217055fa43',
 704             'info_dict': {
 705                 'id': '048195-004_PLUS7-F',
 706                 'ext': 'flv',
 707                 'title': 'X:enius',
 708                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 709                 'upload_date': '20140320',
 710             },
 711             'params': {
 712                 'skip_download': 'Requires rtmpdump'
 713             },
 714             'skip': 'video gone',
 715         },
 716         # francetv embed
 717         {
 718             'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
 719             'info_dict': {
 720                 'id': 'EV_30231',
 721                 'ext': 'mp4',
 722                 'title': 'Alcaline, le concert avec Calogero',
 723                 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
 724                 'upload_date': '20150226',
 725                 'timestamp': 1424989860,
 726                 'duration': 5400,
 727             },
 728             'params': {
 729                 # m3u8 downloads
 730                 'skip_download': True,
 731             },
 732             'expected_warnings': [
 733                 'Forbidden'
 734             ]
 735         },
 736         # Condé Nast embed
 737         {
 738             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 739             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 740             'info_dict': {
 741                 'id': '53501be369702d3275860000',
 742                 'ext': 'mp4',
 743                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 744             }
 745         },
 746         # Dailymotion embed
 747         {
 748             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 749             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 750             'info_dict': {
 751                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 752                 'ext': 'mp4',
 753                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 754                 'description': 'md5:faf028e48a461b8b7fad38f1e104b119',
 755                 'uploader': 'Spi0n',
 756                 'uploader_id': 'xgditw',
 757                 'upload_date': '20140425',
 758                 'timestamp': 1398441542,
 759             },
 760             'add_ie': ['Dailymotion'],
 761         },
 762         # YouTube embed
 763         {
 764             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 765             'info_dict': {
 766                 'id': 'FXRb4ykk4S0',
 767                 'ext': 'mp4',
 768                 'title': 'The NBL Auction 2014',
 769                 'uploader': 'BADMINTON England',
 770                 'uploader_id': 'BADMINTONEvents',
 771                 'upload_date': '20140603',
 772                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 773             },
 774             'add_ie': ['Youtube'],
 775             'params': {
 776                 'skip_download': True,
 777             }
 778         },
 779         # MTVSercices embed
 780         {
 781             'url': 'http://www.vulture.com/2016/06/new-key-peele-sketches-released.html',
 782             'md5': 'ca1aef97695ef2c1d6973256a57e5252',
 783             'info_dict': {
 784                 'id': '769f7ec0-0692-4d62-9b45-0d88074bffc1',
 785                 'ext': 'mp4',
 786                 'title': 'Key and Peele|October 10, 2012|2|203|Liam Neesons - Uncensored',
 787                 'description': 'Two valets share their love for movie star Liam Neesons.',
 788                 'timestamp': 1349922600,
 789                 'upload_date': '20121011',
 790             },
 791         },
 792         # YouTube embed via <data-embed-url="">
 793         {
 794             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 795             'info_dict': {
 796                 'id': '4vAffPZIT44',
 797                 'ext': 'mp4',
 798                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 799                 'uploader': 'Gameloft',
 800                 'uploader_id': 'gameloft',
 801                 'upload_date': '20140828',
 802                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 803             },
 804             'params': {
 805                 'skip_download': True,
 806             }
 807         },
 808         # YouTube <object> embed
 809         {
 810             'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
 811             'md5': '516718101ec834f74318df76259fb3cc',
 812             'info_dict': {
 813                 'id': 'msN87y-iEx0',
 814                 'ext': 'webm',
 815                 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
 816                 'upload_date': '20080526',
 817                 'description': 'md5:0ffc78ea3f01b2e2c247d5f8d1d3c18d',
 818                 'uploader': 'Christopher Sykes',
 819                 'uploader_id': 'ChristopherJSykes',
 820             },
 821             'add_ie': ['Youtube'],
 822         },
 823         # Camtasia studio
 824         {
 825             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 826             'playlist': [{
 827                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 828                 'info_dict': {
 829                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 830                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 831                     'ext': 'flv',
 832                     'duration': 2235.90,
 833                 }
 834             }, {
 835                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 836                 'info_dict': {
 837                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 838                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 839                     'ext': 'flv',
 840                     'duration': 2235.93,
 841                 }
 842             }],
 843             'info_dict': {
 844                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 845             }
 846         },
 847         # Flowplayer
 848         {
 849             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 850             'md5': '9d65602bf31c6e20014319c7d07fba27',
 851             'info_dict': {
 852                 'id': '5123ea6d5e5a7',
 853                 'ext': 'mp4',
 854                 'age_limit': 18,
 855                 'uploader': 'www.handjobhub.com',
 856                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 857             }
 858         },
 859         # Multiple brightcove videos
 860         # https://github.com/rg3/youtube-dl/issues/2283
 861         {
 862             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 863             'info_dict': {
 864                 'id': 'always-never',
 865                 'title': 'Always / Never - The New Yorker',
 866             },
 867             'playlist_count': 3,
 868             'params': {
 869                 'extract_flat': False,
 870                 'skip_download': True,
 871             }
 872         },
 873         # MLB embed
 874         {
 875             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 876             'md5': '96f09a37e44da40dd083e12d9a683327',
 877             'info_dict': {
 878                 'id': '33322633',
 879                 'ext': 'mp4',
 880                 'title': 'Ump changes call to ball',
 881                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 882                 'duration': 48,
 883                 'timestamp': 1401537900,
 884                 'upload_date': '20140531',
 885                 'thumbnail': r're:^https?://.*\.jpg$',
 886             },
 887         },
 888         # Wistia embed
 889         {
 890             'url': 'http://study.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 891             'md5': '1953f3a698ab51cfc948ed3992a0b7ff',
 892             'info_dict': {
 893                 'id': '6e2wtrbdaf',
 894                 'ext': 'mov',
 895                 'title': 'paywall_north-american-exploration-failed-colonies-of-spain-france-england',
 896                 'description': 'a Paywall Videos video from Remilon',
 897                 'duration': 644.072,
 898                 'uploader': 'study.com',
 899                 'timestamp': 1459678540,
 900                 'upload_date': '20160403',
 901                 'filesize': 24687186,
 902             },
 903         },
 904         {
 905             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 906             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 907             'info_dict': {
 908                 'id': 'uxjb0lwrcz',
 909                 'ext': 'mp4',
 910                 'title': 'Conversation about Hexagonal Rails Part 1',
 911                 'description': 'a Martin Fowler video from ThoughtWorks',
 912                 'duration': 1715.0,
 913                 'uploader': 'thoughtworks.wistia.com',
 914                 'timestamp': 1401832161,
 915                 'upload_date': '20140603',
 916             },
 917         },
 918         # Wistia standard embed (async)
 919         {
 920             'url': 'https://www.getdrip.com/university/brennan-dunn-drip-workshop/',
 921             'info_dict': {
 922                 'id': '807fafadvk',
 923                 'ext': 'mp4',
 924                 'title': 'Drip Brennan Dunn Workshop',
 925                 'description': 'a JV Webinars video from getdrip-1',
 926                 'duration': 4986.95,
 927                 'timestamp': 1463607249,
 928                 'upload_date': '20160518',
 929             },
 930             'params': {
 931                 'skip_download': True,
 932             }
 933         },
 934         # Soundcloud embed
 935         {
 936             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 937             'info_dict': {
 938                 'id': '174391317',
 939                 'ext': 'mp3',
 940                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 941                 'uploader': 'Sophos Security',
 942                 'title': 'Chet Chat 171 - Oct 29, 2014',
 943                 'upload_date': '20141029',
 944             }
 945         },
 946         # Soundcloud multiple embeds
 947         {
 948             'url': 'http://www.guitarplayer.com/lessons/1014/legato-workout-one-hour-to-more-fluid-performance---tab/52809',
 949             'info_dict': {
 950                 'id': '52809',
 951                 'title': 'Guitar Essentials: Legato Workout—One-Hour to Fluid Performance  | TAB + AUDIO',
 952             },
 953             'playlist_mincount': 7,
 954         },
 955         # TuneIn station embed
 956         {
 957             'url': 'http://radiocnrv.com/promouvoir-radio-cnrv/',
 958             'info_dict': {
 959                 'id': '204146',
 960                 'ext': 'mp3',
 961                 'title': 'CNRV',
 962                 'location': 'Paris, France',
 963                 'is_live': True,
 964             },
 965             'params': {
 966                 # Live stream
 967                 'skip_download': True,
 968             },
 969         },
 970         # Livestream embed
 971         {
 972             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 973             'info_dict': {
 974                 'id': '67864563',
 975                 'ext': 'flv',
 976                 'upload_date': '20141112',
 977                 'title': 'Rosetta #CometLanding webcast HL 10',
 978             }
 979         },
 980         # Another Livestream embed, without 'new.' in URL
 981         {
 982             'url': 'https://www.freespeech.org/',
 983             'info_dict': {
 984                 'id': '123537347',
 985                 'ext': 'mp4',
 986                 'title': 're:^FSTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 987             },
 988             'params': {
 989                 # Live stream
 990                 'skip_download': True,
 991             },
 992         },
 993         # LazyYT
 994         {
 995             'url': 'https://skiplagged.com/',
 996             'info_dict': {
 997                 'id': 'skiplagged',
 998                 'title': 'Skiplagged: The smart way to find cheap flights',
 999             },
1000             'playlist_mincount': 1,
1001             'add_ie': ['Youtube'],
1002         },
1003         # Cinchcast embed
1004         {
1005             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
1006             'info_dict': {
1007                 'id': '7141703',
1008                 'ext': 'mp3',
1009                 'upload_date': '20141126',
1010                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
1011             }
1012         },
1013         # Cinerama player
1014         {
1015             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
1016             'info_dict': {
1017                 'id': '730m_DandD_1901_512k',
1018                 'ext': 'mp4',
1019                 'uploader': 'www.abc.net.au',
1020                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
1021             }
1022         },
1023         # embedded viddler video
1024         {
1025             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
1026             'info_dict': {
1027                 'id': '4d03aad9',
1028                 'ext': 'mp4',
1029                 'uploader': 'deadspin',
1030                 'title': 'WALL-TO-GORTAT',
1031                 'timestamp': 1422285291,
1032                 'upload_date': '20150126',
1033             },
1034             'add_ie': ['Viddler'],
1035         },
1036         # Libsyn embed
1037         {
1038             'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
1039             'info_dict': {
1040                 'id': '3377616',
1041                 'ext': 'mp3',
1042                 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
1043                 'description': 'md5:601cb790edd05908957dae8aaa866465',
1044                 'upload_date': '20150220',
1045             },
1046             'skip': 'All The Daily Show URLs now redirect to http://www.cc.com/shows/',
1047         },
1048         # jwplayer YouTube
1049         {
1050             'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
1051             'info_dict': {
1052                 'id': 'Mrj4DVp2zeA',
1053                 'ext': 'mp4',
1054                 'upload_date': '20150212',
1055                 'uploader': 'The National Archives UK',
1056                 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
1057                 'uploader_id': 'NationalArchives08',
1058                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
1059             },
1060         },
1061         # jwplayer rtmp
1062         {
1063             'url': 'http://www.suffolk.edu/sjc/',
1064             'info_dict': {
1065                 'id': 'sjclive',
1066                 'ext': 'flv',
1067                 'title': 'Massachusetts Supreme Judicial Court Oral Arguments',
1068                 'uploader': 'www.suffolk.edu',
1069             },
1070             'params': {
1071                 'skip_download': True,
1072             }
1073         },
1074         # Complex jwplayer
1075         {
1076             'url': 'http://www.indiedb.com/games/king-machine/videos',
1077             'info_dict': {
1078                 'id': 'videos',
1079                 'ext': 'mp4',
1080                 'title': 'king machine trailer 1',
1081                 'thumbnail': r're:^https?://.*\.jpg$',
1082             },
1083         },
1084         {
1085             # JWPlayer config passed as variable
1086             'url': 'http://www.txxx.com/videos/3326530/ariele/',
1087             'info_dict': {
1088                 'id': '3326530_hq',
1089                 'ext': 'mp4',
1090                 'title': 'ARIELE | Tube Cup',
1091                 'uploader': 'www.txxx.com',
1092                 'age_limit': 18,
1093             },
1094             'params': {
1095                 'skip_download': True,
1096             }
1097         },
1098         # rtl.nl embed
1099         {
1100             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
1101             'playlist_mincount': 5,
1102             'info_dict': {
1103                 'id': 'aanslagen-kopenhagen',
1104                 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
1105             }
1106         },
1107         # Zapiks embed
1108         {
1109             'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
1110             'info_dict': {
1111                 'id': '118046',
1112                 'ext': 'mp4',
1113                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
1114             }
1115         },
1116         # Kaltura embed (different embed code)
1117         {
1118             'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
1119             'info_dict': {
1120                 'id': '1_a52wc67y',
1121                 'ext': 'flv',
1122                 'upload_date': '20150127',
1123                 'uploader_id': 'PremierMedia',
1124                 'timestamp': int,
1125                 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
1126             },
1127         },
1128         # Kaltura embed with single quotes
1129         {
1130             'url': 'http://fod.infobase.com/p_ViewPlaylist.aspx?AssignmentID=NUN8ZY',
1131             'info_dict': {
1132                 'id': '0_izeg5utt',
1133                 'ext': 'mp4',
1134                 'title': '35871',
1135                 'timestamp': 1355743100,
1136                 'upload_date': '20121217',
1137                 'uploader_id': 'batchUser',
1138             },
1139             'add_ie': ['Kaltura'],
1140         },
1141         {
1142             # Kaltura embedded via quoted entry_id
1143             'url': 'https://www.oreilly.com/ideas/my-cloud-makes-pretty-pictures',
1144             'info_dict': {
1145                 'id': '0_utuok90b',
1146                 'ext': 'mp4',
1147                 'title': '06_matthew_brender_raj_dutt',
1148                 'timestamp': 1466638791,
1149                 'upload_date': '20160622',
1150             },
1151             'add_ie': ['Kaltura'],
1152             'expected_warnings': [
1153                 'Could not send HEAD request'
1154             ],
1155             'params': {
1156                 'skip_download': True,
1157             }
1158         },
1159         {
1160             # Kaltura embedded, some fileExt broken (#11480)
1161             'url': 'http://www.cornell.edu/video/nima-arkani-hamed-standard-models-of-particle-physics',
1162             'info_dict': {
1163                 'id': '1_sgtvehim',
1164                 'ext': 'mp4',
1165                 'title': 'Our "Standard Models" of particle physics and cosmology',
1166                 'description': 'md5:67ea74807b8c4fea92a6f38d6d323861',
1167                 'timestamp': 1321158993,
1168                 'upload_date': '20111113',
1169                 'uploader_id': 'kps1',
1170             },
1171             'add_ie': ['Kaltura'],
1172         },
1173         {
1174             # Kaltura iframe embed
1175             'url': 'http://www.gsd.harvard.edu/event/i-m-pei-a-centennial-celebration/',
1176             'md5': 'ae5ace8eb09dc1a35d03b579a9c2cc44',
1177             'info_dict': {
1178                 'id': '0_f2cfbpwy',
1179                 'ext': 'mp4',
1180                 'title': 'I. M. Pei: A Centennial Celebration',
1181                 'description': 'md5:1db8f40c69edc46ca180ba30c567f37c',
1182                 'upload_date': '20170403',
1183                 'uploader_id': 'batchUser',
1184                 'timestamp': 1491232186,
1185             },
1186             'add_ie': ['Kaltura'],
1187         },
1188         # EaglePlatform embed (generic URL)
1189         {
1190             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
1191             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
1192             'info_dict': {
1193                 'id': '227304',
1194                 'ext': 'mp4',
1195                 'title': 'Навальный вышел на свободу',
1196                 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
1197                 'thumbnail': r're:^https?://.*\.jpg$',
1198                 'duration': 87,
1199                 'view_count': int,
1200                 'age_limit': 0,
1201             },
1202             'params': {
1203                 'skip_download': True,
1204             },
1205         },
1206         # referrer protected EaglePlatform embed
1207         {
1208             'url': 'https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/',
1209             'info_dict': {
1210                 'id': '582306',
1211                 'ext': 'mp4',
1212                 'title': 'Стас Намин: «Мы нарушили девственность Кремля»',
1213                 'thumbnail': r're:^https?://.*\.jpg$',
1214                 'duration': 3382,
1215                 'view_count': int,
1216             },
1217             'params': {
1218                 'skip_download': True,
1219             },
1220         },
1221         # ClipYou (EaglePlatform) embed (custom URL)
1222         {
1223             'url': 'http://muz-tv.ru/play/7129/',
1224             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
1225             'info_dict': {
1226                 'id': '12820',
1227                 'ext': 'mp4',
1228                 'title': "'O Sole Mio",
1229                 'thumbnail': r're:^https?://.*\.jpg$',
1230                 'duration': 216,
1231                 'view_count': int,
1232             },
1233             'params': {
1234                 'skip_download': True,
1235             },
1236         },
1237         # Pladform embed
1238         {
1239             'url': 'http://muz-tv.ru/kinozal/view/7400/',
1240             'info_dict': {
1241                 'id': '100183293',
1242                 'ext': 'mp4',
1243                 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
1244                 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
1245                 'thumbnail': r're:^https?://.*\.jpg$',
1246                 'duration': 694,
1247                 'age_limit': 0,
1248             },
1249         },
1250         # Playwire embed
1251         {
1252             'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
1253             'info_dict': {
1254                 'id': '3519514',
1255                 'ext': 'mp4',
1256                 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
1257                 'thumbnail': r're:^https?://.*\.png$',
1258                 'duration': 45.115,
1259             },
1260         },
1261         # 5min embed
1262         {
1263             'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
1264             'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
1265             'info_dict': {
1266                 'id': '518726732',
1267                 'ext': 'mp4',
1268                 'title': 'Facebook Creates "On This Day" | Crunch Report',
1269             },
1270         },
1271         # SVT embed
1272         {
1273             'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
1274             'info_dict': {
1275                 'id': '2900353',
1276                 'ext': 'flv',
1277                 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
1278                 'duration': 27,
1279                 'age_limit': 0,
1280             },
1281         },
1282         # Crooks and Liars embed
1283         {
1284             'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
1285             'info_dict': {
1286                 'id': '8RUoRhRi',
1287                 'ext': 'mp4',
1288                 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
1289                 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
1290                 'timestamp': 1428207000,
1291                 'upload_date': '20150405',
1292                 'uploader': 'Heather',
1293             },
1294         },
1295         # Crooks and Liars external embed
1296         {
1297             'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
1298             'info_dict': {
1299                 'id': 'MTE3MjUtMzQ2MzA',
1300                 'ext': 'mp4',
1301                 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
1302                 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
1303                 'timestamp': 1265032391,
1304                 'upload_date': '20100201',
1305                 'uploader': 'Heather',
1306             },
1307         },
1308         # NBC Sports vplayer embed
1309         {
1310             'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
1311             'info_dict': {
1312                 'id': 'ln7x1qSThw4k',
1313                 'ext': 'flv',
1314                 'title': "PFT Live: New leader in the 'new-look' defense",
1315                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
1316                 'uploader': 'NBCU-SPORTS',
1317                 'upload_date': '20140107',
1318                 'timestamp': 1389118457,
1319             },
1320         },
1321         # NBC News embed
1322         {
1323             'url': 'http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html',
1324             'md5': '1aa589c675898ae6d37a17913cf68d66',
1325             'info_dict': {
1326                 'id': '701714499682',
1327                 'ext': 'mp4',
1328                 'title': 'PREVIEW: On Assignment: David Letterman',
1329                 '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.',
1330             },
1331         },
1332         # UDN embed
1333         {
1334             'url': 'https://video.udn.com/news/300346',
1335             'md5': 'fd2060e988c326991037b9aff9df21a6',
1336             'info_dict': {
1337                 'id': '300346',
1338                 'ext': 'mp4',
1339                 'title': '中一中男師變性 全校師生力挺',
1340                 'thumbnail': r're:^https?://.*\.jpg$',
1341             },
1342             'params': {
1343                 # m3u8 download
1344                 'skip_download': True,
1345             },
1346         },
1347         # Ooyala embed
1348         {
1349             'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
1350             'info_dict': {
1351                 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
1352                 'ext': 'mp4',
1353                 'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.',
1354                 'title': 'This is what separates the Excel masters from the wannabes',
1355                 'duration': 191.933,
1356             },
1357             'params': {
1358                 # m3u8 downloads
1359                 'skip_download': True,
1360             }
1361         },
1362         # Brightcove URL in single quotes
1363         {
1364             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
1365             'md5': '4ae374f1f8b91c889c4b9203c8c752af',
1366             'info_dict': {
1367                 'id': '4255764656001',
1368                 'ext': 'mp4',
1369                 'title': 'SN Presents: Russell Martin, World Citizen',
1370                 '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.',
1371                 'uploader': 'Rogers Sportsnet',
1372                 'uploader_id': '1704050871',
1373                 'upload_date': '20150525',
1374                 'timestamp': 1432570283,
1375             },
1376         },
1377         # Dailymotion Cloud video
1378         {
1379             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
1380             'md5': 'dcaf23ad0c67a256f4278bce6e0bae38',
1381             'info_dict': {
1382                 'id': 'x2uy8t3',
1383                 'ext': 'mp4',
1384                 'title': 'Sauvons les abeilles ! - Le débat',
1385                 'description': 'md5:d9082128b1c5277987825d684939ca26',
1386                 'thumbnail': r're:^https?://.*\.jpe?g$',
1387                 'timestamp': 1434970506,
1388                 'upload_date': '20150622',
1389                 'uploader': 'Public Sénat',
1390                 'uploader_id': 'xa9gza',
1391             }
1392         },
1393         # OnionStudios embed
1394         {
1395             'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
1396             'info_dict': {
1397                 'id': '2855',
1398                 'ext': 'mp4',
1399                 'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
1400                 'thumbnail': r're:^https?://.*\.jpe?g$',
1401                 'uploader': 'ClickHole',
1402                 'uploader_id': 'clickhole',
1403             }
1404         },
1405         # SnagFilms embed
1406         {
1407             'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
1408             'info_dict': {
1409                 'id': '74849a00-85a9-11e1-9660-123139220831',
1410                 'ext': 'mp4',
1411                 'title': '#whilewewatch',
1412             }
1413         },
1414         # AdobeTVVideo embed
1415         {
1416             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
1417             'md5': '43662b577c018ad707a63766462b1e87',
1418             'info_dict': {
1419                 'id': '2456',
1420                 'ext': 'mp4',
1421                 'title': 'New experience with Acrobat DC',
1422                 'description': 'New experience with Acrobat DC',
1423                 'duration': 248.667,
1424             },
1425         },
1426         # BrightcoveInPageEmbed embed
1427         {
1428             'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
1429             'info_dict': {
1430                 'id': '4238694884001',
1431                 'ext': 'flv',
1432                 'title': 'Tabletop: Dread, Last Thoughts',
1433                 'description': 'Tabletop: Dread, Last Thoughts',
1434                 'duration': 51690,
1435             },
1436         },
1437         # Brightcove embed, with no valid 'renditions' but valid 'IOSRenditions'
1438         # This video can't be played in browsers if Flash disabled and UA set to iPhone, which is actually a false alarm
1439         {
1440             'url': 'https://dl.dropboxusercontent.com/u/29092637/interview.html',
1441             'info_dict': {
1442                 'id': '4785848093001',
1443                 'ext': 'mp4',
1444                 'title': 'The Cardinal Pell Interview',
1445                 'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ',
1446                 'uploader': 'GlobeCast Australia - GlobeStream',
1447                 'uploader_id': '2733773828001',
1448                 'upload_date': '20160304',
1449                 'timestamp': 1457083087,
1450             },
1451             'params': {
1452                 # m3u8 downloads
1453                 'skip_download': True,
1454             },
1455         },
1456         {
1457             # Brightcove embed with whitespace around attribute names
1458             'url': 'http://www.stack.com/video/3167554373001/learn-to-hit-open-three-pointers-with-damian-lillard-s-baseline-drift-drill',
1459             'info_dict': {
1460                 'id': '3167554373001',
1461                 'ext': 'mp4',
1462                 'title': "Learn to Hit Open Three-Pointers With Damian Lillard's Baseline Drift Drill",
1463                 'description': 'md5:57bacb0e0f29349de4972bfda3191713',
1464                 'uploader_id': '1079349493',
1465                 'upload_date': '20140207',
1466                 'timestamp': 1391810548,
1467             },
1468             'params': {
1469                 'skip_download': True,
1470             },
1471         },
1472         # Another form of arte.tv embed
1473         {
1474             'url': 'http://www.tv-replay.fr/redirection/09-04-16/arte-reportage-arte-11508975.html',
1475             'md5': '850bfe45417ddf221288c88a0cffe2e2',
1476             'info_dict': {
1477                 'id': '030273-562_PLUS7-F',
1478                 'ext': 'mp4',
1479                 'title': 'ARTE Reportage - Nulle part, en France',
1480                 'description': 'md5:e3a0e8868ed7303ed509b9e3af2b870d',
1481                 'upload_date': '20160409',
1482             },
1483         },
1484         # LiveLeak embed
1485         {
1486             'url': 'http://www.wykop.pl/link/3088787/',
1487             'md5': 'ace83b9ed19b21f68e1b50e844fdf95d',
1488             'info_dict': {
1489                 'id': '874_1459135191',
1490                 'ext': 'mp4',
1491                 'title': 'Man shows poor quality of new apartment building',
1492                 'description': 'The wall is like a sand pile.',
1493                 'uploader': 'Lake8737',
1494             }
1495         },
1496         # Duplicated embedded video URLs
1497         {
1498             'url': 'http://www.hudl.com/athlete/2538180/highlights/149298443',
1499             'info_dict': {
1500                 'id': '149298443_480_16c25b74_2',
1501                 'ext': 'mp4',
1502                 'title': 'vs. Blue Orange Spring Game',
1503                 'uploader': 'www.hudl.com',
1504             },
1505         },
1506         # twitter:player:stream embed
1507         {
1508             'url': 'http://www.rtl.be/info/video/589263.aspx?CategoryID=288',
1509             'info_dict': {
1510                 'id': 'master',
1511                 'ext': 'mp4',
1512                 'title': 'Une nouvelle espèce de dinosaure découverte en Argentine',
1513                 'uploader': 'www.rtl.be',
1514             },
1515             'params': {
1516                 # m3u8 downloads
1517                 'skip_download': True,
1518             },
1519         },
1520         # twitter:player embed
1521         {
1522             'url': 'http://www.theatlantic.com/video/index/484130/what-do-black-holes-sound-like/',
1523             'md5': 'a3e0df96369831de324f0778e126653c',
1524             'info_dict': {
1525                 'id': '4909620399001',
1526                 'ext': 'mp4',
1527                 'title': 'What Do Black Holes Sound Like?',
1528                 'description': 'what do black holes sound like',
1529                 'upload_date': '20160524',
1530                 'uploader_id': '29913724001',
1531                 'timestamp': 1464107587,
1532                 'uploader': 'TheAtlantic',
1533             },
1534             'add_ie': ['BrightcoveLegacy'],
1535         },
1536         # Facebook <iframe> embed
1537         {
1538             'url': 'https://www.hostblogger.de/blog/archives/6181-Auto-jagt-Betonmischer.html',
1539             'md5': 'fbcde74f534176ecb015849146dd3aee',
1540             'info_dict': {
1541                 'id': '599637780109885',
1542                 'ext': 'mp4',
1543                 'title': 'Facebook video #599637780109885',
1544             },
1545         },
1546         # Facebook <iframe> embed, plugin video
1547         {
1548             'url': 'http://5pillarsuk.com/2017/06/07/tariq-ramadan-disagrees-with-pr-exercise-by-imams-refusing-funeral-prayers-for-london-attackers/',
1549             'info_dict': {
1550                 'id': '1754168231264132',
1551                 'ext': 'mp4',
1552                 'title': 'About the Imams and Religious leaders refusing to perform funeral prayers for...',
1553                 'uploader': 'Tariq Ramadan (official)',
1554                 'timestamp': 1496758379,
1555                 'upload_date': '20170606',
1556             },
1557             'params': {
1558                 'skip_download': True,
1559             },
1560         },
1561         # Facebook API embed
1562         {
1563             'url': 'http://www.lothype.com/blue-stars-2016-preview-standstill-full-show/',
1564             'md5': 'a47372ee61b39a7b90287094d447d94e',
1565             'info_dict': {
1566                 'id': '10153467542406923',
1567                 'ext': 'mp4',
1568                 'title': 'Facebook video #10153467542406923',
1569             },
1570         },
1571         # Wordpress "YouTube Video Importer" plugin
1572         {
1573             'url': 'http://www.lothype.com/blue-devils-drumline-stanford-lot-2016/',
1574             'md5': 'd16797741b560b485194eddda8121b48',
1575             'info_dict': {
1576                 'id': 'HNTXWDXV9Is',
1577                 'ext': 'mp4',
1578                 'title': 'Blue Devils Drumline Stanford lot 2016',
1579                 'upload_date': '20160627',
1580                 'uploader_id': 'GENOCIDE8GENERAL10',
1581                 'uploader': 'cylus cyrus',
1582             },
1583         },
1584         {
1585             # video stored on custom kaltura server
1586             'url': 'http://www.expansion.com/multimedia/videos.html?media=EQcM30NHIPv',
1587             'md5': '537617d06e64dfed891fa1593c4b30cc',
1588             'info_dict': {
1589                 'id': '0_1iotm5bh',
1590                 'ext': 'mp4',
1591                 'title': 'Elecciones británicas: 5 lecciones para Rajoy',
1592                 'description': 'md5:435a89d68b9760b92ce67ed227055f16',
1593                 'uploader_id': 'videos.expansion@el-mundo.net',
1594                 'upload_date': '20150429',
1595                 'timestamp': 1430303472,
1596             },
1597             'add_ie': ['Kaltura'],
1598         },
1599         {
1600             # Non-standard Vimeo embed
1601             'url': 'https://openclassrooms.com/courses/understanding-the-web',
1602             'md5': '64d86f1c7d369afd9a78b38cbb88d80a',
1603             'info_dict': {
1604                 'id': '148867247',
1605                 'ext': 'mp4',
1606                 'title': 'Understanding the web - Teaser',
1607                 'description': 'This is "Understanding the web - Teaser" by openclassrooms on Vimeo, the home for high quality videos and the people who love them.',
1608                 'upload_date': '20151214',
1609                 'uploader': 'OpenClassrooms',
1610                 'uploader_id': 'openclassrooms',
1611             },
1612             'add_ie': ['Vimeo'],
1613         },
1614         {
1615             # generic vimeo embed that requires original URL passed as Referer
1616             'url': 'http://racing4everyone.eu/2016/07/30/formula-1-2016-round12-germany/',
1617             'only_matching': True,
1618         },
1619         {
1620             'url': 'https://support.arkena.com/display/PLAY/Ways+to+embed+your+video',
1621             'md5': 'b96f2f71b359a8ecd05ce4e1daa72365',
1622             'info_dict': {
1623                 'id': 'b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe',
1624                 'ext': 'mp4',
1625                 'title': 'Big Buck Bunny',
1626                 'description': 'Royalty free test video',
1627                 'timestamp': 1432816365,
1628                 'upload_date': '20150528',
1629                 'is_live': False,
1630             },
1631             'params': {
1632                 'skip_download': True,
1633             },
1634             'add_ie': [ArkenaIE.ie_key()],
1635         },
1636         {
1637             '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/',
1638             'info_dict': {
1639                 'id': '1c7141f46c',
1640                 'ext': 'mp4',
1641                 'title': 'НА КОСЪМ ОТ ВЗРИВ: Изтичане на газ на бензиностанция в Пловдив',
1642             },
1643             'params': {
1644                 'skip_download': True,
1645             },
1646             'add_ie': [Vbox7IE.ie_key()],
1647         },
1648         {
1649             # DBTV embeds
1650             'url': 'http://www.dagbladet.no/2016/02/23/nyheter/nordlys/ski/troms/ver/43254897/',
1651             'info_dict': {
1652                 'id': '43254897',
1653                 'title': 'Etter ett års planlegging, klaffet endelig alt: - Jeg måtte ta en liten dans',
1654             },
1655             'playlist_mincount': 3,
1656         },
1657         {
1658             # Videa embeds
1659             'url': 'http://forum.dvdtalk.com/movie-talk/623756-deleted-magic-star-wars-ot-deleted-alt-scenes-docu-style.html',
1660             'info_dict': {
1661                 'id': '623756-deleted-magic-star-wars-ot-deleted-alt-scenes-docu-style',
1662                 'title': 'Deleted Magic - Star Wars: OT Deleted / Alt. Scenes Docu. Style - DVD Talk Forum',
1663             },
1664             'playlist_mincount': 2,
1665         },
1666         {
1667             # 20 minuten embed
1668             'url': 'http://www.20min.ch/schweiz/news/story/So-kommen-Sie-bei-Eis-und-Schnee-sicher-an-27032552',
1669             'info_dict': {
1670                 'id': '523629',
1671                 'ext': 'mp4',
1672                 'title': 'So kommen Sie bei Eis und Schnee sicher an',
1673                 'description': 'md5:117c212f64b25e3d95747e5276863f7d',
1674             },
1675             'params': {
1676                 'skip_download': True,
1677             },
1678             'add_ie': [TwentyMinutenIE.ie_key()],
1679         },
1680         {
1681             # VideoPress embed
1682             'url': 'https://en.support.wordpress.com/videopress/',
1683             'info_dict': {
1684                 'id': 'OcobLTqC',
1685                 'ext': 'm4v',
1686                 'title': 'IMG_5786',
1687                 'timestamp': 1435711927,
1688                 'upload_date': '20150701',
1689             },
1690             'params': {
1691                 'skip_download': True,
1692             },
1693             'add_ie': [VideoPressIE.ie_key()],
1694         },
1695         {
1696             # Rutube embed
1697             'url': 'http://magazzino.friday.ru/videos/vipuski/kazan-2',
1698             'info_dict': {
1699                 'id': '9b3d5bee0a8740bf70dfd29d3ea43541',
1700                 'ext': 'flv',
1701                 'title': 'Магаззино: Казань 2',
1702                 'description': 'md5:99bccdfac2269f0e8fdbc4bbc9db184a',
1703                 'uploader': 'Магаззино',
1704                 'upload_date': '20170228',
1705                 'uploader_id': '996642',
1706             },
1707             'params': {
1708                 'skip_download': True,
1709             },
1710             'add_ie': [RutubeIE.ie_key()],
1711         },
1712         {
1713             # ThePlatform embedded with whitespaces in URLs
1714             'url': 'http://www.golfchannel.com/topics/shows/golftalkcentral.htm',
1715             'only_matching': True,
1716         },
1717         {
1718             # Senate ISVP iframe https
1719             'url': 'https://www.hsgac.senate.gov/hearings/canadas-fast-track-refugee-plan-unanswered-questions-and-implications-for-us-national-security',
1720             'md5': 'fb8c70b0b515e5037981a2492099aab8',
1721             'info_dict': {
1722                 'id': 'govtaff020316',
1723                 'ext': 'mp4',
1724                 'title': 'Integrated Senate Video Player',
1725             },
1726             'add_ie': [SenateISVPIE.ie_key()],
1727         },
1728         {
1729             # Limelight embeds (1 channel embed + 4 media embeds)
1730             'url': 'http://www.sedona.com/FacilitatorTraining2017',
1731             'info_dict': {
1732                 'id': 'FacilitatorTraining2017',
1733                 'title': 'Facilitator Training 2017',
1734             },
1735             'playlist_mincount': 5,
1736         },
1737         {
1738             'url': 'http://kron4.com/2017/04/28/standoff-with-walnut-creek-murder-suspect-ends-with-arrest/',
1739             'info_dict': {
1740                 'id': 'standoff-with-walnut-creek-murder-suspect-ends-with-arrest',
1741                 'title': 'Standoff with Walnut Creek murder suspect ends',
1742                 'description': 'md5:3ccc48a60fc9441eeccfc9c469ebf788',
1743             },
1744             'playlist_mincount': 4,
1745         },
1746         {
1747             # WashingtonPost embed
1748             'url': 'http://www.vanityfair.com/hollywood/2017/04/donald-trump-tv-pitches',
1749             'info_dict': {
1750                 'id': '8caf6e88-d0ec-11e5-90d3-34c2c42653ac',
1751                 'ext': 'mp4',
1752                 'title': "No one has seen the drama series based on Trump's life \u2014 until now",
1753                 'description': 'Donald Trump wanted a weekly TV drama based on his life. It never aired. But The Washington Post recently obtained a scene from the pilot script — and enlisted actors.',
1754                 'timestamp': 1455216756,
1755                 'uploader': 'The Washington Post',
1756                 'upload_date': '20160211',
1757             },
1758             'add_ie': [WashingtonPostIE.ie_key()],
1759         },
1760         {
1761             # Mediaset embed
1762             'url': 'http://www.tgcom24.mediaset.it/politica/serracchiani-voglio-vivere-in-una-societa-aperta-reazioni-sproporzionate-_3071354-201702a.shtml',
1763             'info_dict': {
1764                 'id': '720642',
1765                 'ext': 'mp4',
1766                 'title': 'Serracchiani: "Voglio vivere in una società aperta, con tutela del patto di fiducia"',
1767             },
1768             'params': {
1769                 'skip_download': True,
1770             },
1771             'add_ie': [MediasetIE.ie_key()],
1772         },
1773         # {
1774         #     # TODO: find another test
1775         #     # http://schema.org/VideoObject
1776         #     'url': 'https://flipagram.com/f/nyvTSJMKId',
1777         #     'md5': '888dcf08b7ea671381f00fab74692755',
1778         #     'info_dict': {
1779         #         'id': 'nyvTSJMKId',
1780         #         'ext': 'mp4',
1781         #         'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction',
1782         #         'description': '#love for cats.',
1783         #         'timestamp': 1461244995,
1784         #         'upload_date': '20160421',
1785         #     },
1786         #     'params': {
1787         #         'force_generic_extractor': True,
1788         #     },
1789         # }
1790     ]
1791
1792     def report_following_redirect(self, new_url):
1793         """Report information extraction."""
1794         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
1795
1796     def _extract_rss(self, url, video_id, doc):
1797         playlist_title = doc.find('./channel/title').text
1798         playlist_desc_el = doc.find('./channel/description')
1799         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
1800
1801         entries = []
1802         for it in doc.findall('./channel/item'):
1803             next_url = xpath_text(it, 'link', fatal=False)
1804             if not next_url:
1805                 enclosure_nodes = it.findall('./enclosure')
1806                 for e in enclosure_nodes:
1807                     next_url = e.attrib.get('url')
1808                     if next_url:
1809                         break
1810
1811             if not next_url:
1812                 continue
1813
1814             entries.append({
1815                 '_type': 'url_transparent',
1816                 'url': next_url,
1817                 'title': it.find('title').text,
1818             })
1819
1820         return {
1821             '_type': 'playlist',
1822             'id': url,
1823             'title': playlist_title,
1824             'description': playlist_desc,
1825             'entries': entries,
1826         }
1827
1828     def _extract_camtasia(self, url, video_id, webpage):
1829         """ Returns None if no camtasia video can be found. """
1830
1831         camtasia_cfg = self._search_regex(
1832             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
1833             webpage, 'camtasia configuration file', default=None)
1834         if camtasia_cfg is None:
1835             return None
1836
1837         title = self._html_search_meta('DC.title', webpage, fatal=True)
1838
1839         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
1840         camtasia_cfg = self._download_xml(
1841             camtasia_url, video_id,
1842             note='Downloading camtasia configuration',
1843             errnote='Failed to download camtasia configuration')
1844         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
1845
1846         entries = []
1847         for n in fileset_node.getchildren():
1848             url_n = n.find('./uri')
1849             if url_n is None:
1850                 continue
1851
1852             entries.append({
1853                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
1854                 'title': '%s - %s' % (title, n.tag),
1855                 'url': compat_urlparse.urljoin(url, url_n.text),
1856                 'duration': float_or_none(n.find('./duration').text),
1857             })
1858
1859         return {
1860             '_type': 'playlist',
1861             'entries': entries,
1862             'title': title,
1863         }
1864
1865     def _real_extract(self, url):
1866         if url.startswith('//'):
1867             return {
1868                 '_type': 'url',
1869                 'url': self.http_scheme() + url,
1870             }
1871
1872         parsed_url = compat_urlparse.urlparse(url)
1873         if not parsed_url.scheme:
1874             default_search = self._downloader.params.get('default_search')
1875             if default_search is None:
1876                 default_search = 'fixup_error'
1877
1878             if default_search in ('auto', 'auto_warning', 'fixup_error'):
1879                 if '/' in url:
1880                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
1881                     return self.url_result('http://' + url)
1882                 elif default_search != 'fixup_error':
1883                     if default_search == 'auto_warning':
1884                         if re.match(r'^(?:url|URL)$', url):
1885                             raise ExtractorError(
1886                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
1887                                 expected=True)
1888                         else:
1889                             self._downloader.report_warning(
1890                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
1891                     return self.url_result('ytsearch:' + url)
1892
1893             if default_search in ('error', 'fixup_error'):
1894                 raise ExtractorError(
1895                     '%r is not a valid URL. '
1896                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
1897                     % (url, url), expected=True)
1898             else:
1899                 if ':' not in default_search:
1900                     default_search += ':'
1901                 return self.url_result(default_search + url)
1902
1903         url, smuggled_data = unsmuggle_url(url)
1904         force_videoid = None
1905         is_intentional = smuggled_data and smuggled_data.get('to_generic')
1906         if smuggled_data and 'force_videoid' in smuggled_data:
1907             force_videoid = smuggled_data['force_videoid']
1908             video_id = force_videoid
1909         else:
1910             video_id = self._generic_id(url)
1911
1912         self.to_screen('%s: Requesting header' % video_id)
1913
1914         head_req = HEADRequest(url)
1915         head_response = self._request_webpage(
1916             head_req, video_id,
1917             note=False, errnote='Could not send HEAD request to %s' % url,
1918             fatal=False)
1919
1920         if head_response is not False:
1921             # Check for redirect
1922             new_url = head_response.geturl()
1923             if url != new_url:
1924                 self.report_following_redirect(new_url)
1925                 if force_videoid:
1926                     new_url = smuggle_url(
1927                         new_url, {'force_videoid': force_videoid})
1928                 return self.url_result(new_url)
1929
1930         full_response = None
1931         if head_response is False:
1932             request = sanitized_Request(url)
1933             request.add_header('Accept-Encoding', '*')
1934             full_response = self._request_webpage(request, video_id)
1935             head_response = full_response
1936
1937         info_dict = {
1938             'id': video_id,
1939             'title': self._generic_title(url),
1940             'upload_date': unified_strdate(head_response.headers.get('Last-Modified'))
1941         }
1942
1943         # Check for direct link to a video
1944         content_type = head_response.headers.get('Content-Type', '').lower()
1945         m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>[^;\s]+)', content_type)
1946         if m:
1947             format_id = compat_str(m.group('format_id'))
1948             if format_id.endswith('mpegurl'):
1949                 formats = self._extract_m3u8_formats(url, video_id, 'mp4')
1950             elif format_id == 'f4m':
1951                 formats = self._extract_f4m_formats(url, video_id)
1952             else:
1953                 formats = [{
1954                     'format_id': format_id,
1955                     'url': url,
1956                     'vcodec': 'none' if m.group('type') == 'audio' else None
1957                 }]
1958                 info_dict['direct'] = True
1959             self._sort_formats(formats)
1960             info_dict['formats'] = formats
1961             return info_dict
1962
1963         if not self._downloader.params.get('test', False) and not is_intentional:
1964             force = self._downloader.params.get('force_generic_extractor', False)
1965             self._downloader.report_warning(
1966                 '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
1967
1968         if not full_response:
1969             request = sanitized_Request(url)
1970             # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
1971             # making it impossible to download only chunk of the file (yet we need only 512kB to
1972             # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
1973             # that will always result in downloading the whole file that is not desirable.
1974             # Therefore for extraction pass we have to override Accept-Encoding to any in order
1975             # to accept raw bytes and being able to download only a chunk.
1976             # It may probably better to solve this by checking Content-Type for application/octet-stream
1977             # after HEAD request finishes, but not sure if we can rely on this.
1978             request.add_header('Accept-Encoding', '*')
1979             full_response = self._request_webpage(request, video_id)
1980
1981         first_bytes = full_response.read(512)
1982
1983         # Is it an M3U playlist?
1984         if first_bytes.startswith(b'#EXTM3U'):
1985             info_dict['formats'] = self._extract_m3u8_formats(url, video_id, 'mp4')
1986             self._sort_formats(info_dict['formats'])
1987             return info_dict
1988
1989         # Maybe it's a direct link to a video?
1990         # Be careful not to download the whole thing!
1991         if not is_html(first_bytes):
1992             self._downloader.report_warning(
1993                 'URL could be a direct video link, returning it as such.')
1994             info_dict.update({
1995                 'direct': True,
1996                 'url': url,
1997             })
1998             return info_dict
1999
2000         webpage = self._webpage_read_content(
2001             full_response, url, video_id, prefix=first_bytes)
2002
2003         self.report_extraction(video_id)
2004
2005         # Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest?
2006         try:
2007             doc = compat_etree_fromstring(webpage.encode('utf-8'))
2008             if doc.tag == 'rss':
2009                 return self._extract_rss(url, video_id, doc)
2010             elif doc.tag == 'SmoothStreamingMedia':
2011                 info_dict['formats'] = self._parse_ism_formats(doc, url)
2012                 self._sort_formats(info_dict['formats'])
2013                 return info_dict
2014             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
2015                 smil = self._parse_smil(doc, url, video_id)
2016                 self._sort_formats(smil['formats'])
2017                 return smil
2018             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
2019                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
2020             elif re.match(r'(?i)^(?:{[^}]+})?MPD$', doc.tag):
2021                 info_dict['formats'] = self._parse_mpd_formats(
2022                     doc, video_id,
2023                     mpd_base_url=full_response.geturl().rpartition('/')[0],
2024                     mpd_url=url)
2025                 self._sort_formats(info_dict['formats'])
2026                 return info_dict
2027             elif re.match(r'^{http://ns\.adobe\.com/f4m/[12]\.0}manifest$', doc.tag):
2028                 info_dict['formats'] = self._parse_f4m_formats(doc, url, video_id)
2029                 self._sort_formats(info_dict['formats'])
2030                 return info_dict
2031         except compat_xml_parse_error:
2032             pass
2033
2034         # Is it a Camtasia project?
2035         camtasia_res = self._extract_camtasia(url, video_id, webpage)
2036         if camtasia_res is not None:
2037             return camtasia_res
2038
2039         # Sometimes embedded video player is hidden behind percent encoding
2040         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
2041         # Unescaping the whole page allows to handle those cases in a generic way
2042         webpage = compat_urllib_parse_unquote(webpage)
2043
2044         # it's tempting to parse this further, but you would
2045         # have to take into account all the variations like
2046         #   Video Title - Site Name
2047         #   Site Name | Video Title
2048         #   Video Title - Tagline | Site Name
2049         # and so on and so forth; it's just not practical
2050         video_title = self._og_search_title(
2051             webpage, default=None) or self._html_search_regex(
2052             r'(?s)<title>(.*?)</title>', webpage, 'video title',
2053             default='video')
2054
2055         # Try to detect age limit automatically
2056         age_limit = self._rta_search(webpage)
2057         # And then there are the jokers who advertise that they use RTA,
2058         # but actually don't.
2059         AGE_LIMIT_MARKERS = [
2060             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
2061         ]
2062         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
2063             age_limit = 18
2064
2065         # video uploader is domain name
2066         video_uploader = self._search_regex(
2067             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
2068
2069         video_description = self._og_search_description(webpage, default=None)
2070         video_thumbnail = self._og_search_thumbnail(webpage, default=None)
2071
2072         info_dict.update({
2073             'title': video_title,
2074             'description': video_description,
2075             'thumbnail': video_thumbnail,
2076             'age_limit': age_limit,
2077         })
2078
2079         # Look for Brightcove Legacy Studio embeds
2080         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
2081         if bc_urls:
2082             entries = [{
2083                 '_type': 'url',
2084                 'url': smuggle_url(bc_url, {'Referer': url}),
2085                 'ie_key': 'BrightcoveLegacy'
2086             } for bc_url in bc_urls]
2087
2088             return {
2089                 '_type': 'playlist',
2090                 'title': video_title,
2091                 'id': video_id,
2092                 'entries': entries,
2093             }
2094
2095         # Look for Brightcove New Studio embeds
2096         bc_urls = BrightcoveNewIE._extract_urls(self, webpage)
2097         if bc_urls:
2098             return self.playlist_from_matches(bc_urls, video_id, video_title, ie='BrightcoveNew')
2099
2100         # Look for ThePlatform embeds
2101         tp_urls = ThePlatformIE._extract_urls(webpage)
2102         if tp_urls:
2103             return self.playlist_from_matches(tp_urls, video_id, video_title, ie='ThePlatform')
2104
2105         # Look for Vessel embeds
2106         vessel_urls = VesselIE._extract_urls(webpage)
2107         if vessel_urls:
2108             return self.playlist_from_matches(vessel_urls, video_id, video_title, ie=VesselIE.ie_key())
2109
2110         # Look for embedded rtl.nl player
2111         matches = re.findall(
2112             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
2113             webpage)
2114         if matches:
2115             return self.playlist_from_matches(matches, video_id, video_title, ie='RtlNl')
2116
2117         vimeo_urls = VimeoIE._extract_urls(url, webpage)
2118         if vimeo_urls:
2119             return self.playlist_from_matches(vimeo_urls, video_id, video_title, ie=VimeoIE.ie_key())
2120
2121         vid_me_embed_url = self._search_regex(
2122             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
2123             webpage, 'vid.me embed', default=None)
2124         if vid_me_embed_url is not None:
2125             return self.url_result(vid_me_embed_url, 'Vidme')
2126
2127         # Look for embedded YouTube player
2128         matches = re.findall(r'''(?x)
2129             (?:
2130                 <iframe[^>]+?src=|
2131                 data-video-url=|
2132                 <embed[^>]+?src=|
2133                 embedSWF\(?:\s*|
2134                 <object[^>]+data=|
2135                 new\s+SWFObject\(
2136             )
2137             (["\'])
2138                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2139                 (?:embed|v|p)/.+?)
2140             \1''', webpage)
2141         if matches:
2142             return self.playlist_from_matches(
2143                 matches, video_id, video_title, lambda m: unescapeHTML(m[1]))
2144
2145         # Look for lazyYT YouTube embed
2146         matches = re.findall(
2147             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
2148         if matches:
2149             return self.playlist_from_matches(matches, video_id, video_title, lambda m: unescapeHTML(m))
2150
2151         # Look for Wordpress "YouTube Video Importer" plugin
2152         matches = re.findall(r'''(?x)<div[^>]+
2153             class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2154             data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2155         if matches:
2156             return self.playlist_from_matches(matches, video_id, video_title, lambda m: m[-1])
2157
2158         matches = DailymotionIE._extract_urls(webpage)
2159         if matches:
2160             return self.playlist_from_matches(matches, video_id, video_title)
2161
2162         # Look for embedded Dailymotion playlist player (#3822)
2163         m = re.search(
2164             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
2165         if m:
2166             playlists = re.findall(
2167                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
2168             if playlists:
2169                 return self.playlist_from_matches(
2170                     playlists, video_id, video_title, lambda p: '//dailymotion.com/playlist/%s' % p)
2171
2172         # Look for embedded Wistia player
2173         wistia_url = WistiaIE._extract_url(webpage)
2174         if wistia_url:
2175             return {
2176                 '_type': 'url_transparent',
2177                 'url': self._proto_relative_url(wistia_url),
2178                 'ie_key': WistiaIE.ie_key(),
2179                 'uploader': video_uploader,
2180             }
2181
2182         # Look for SVT player
2183         svt_url = SVTIE._extract_url(webpage)
2184         if svt_url:
2185             return self.url_result(svt_url, 'SVT')
2186
2187         # Look for Bandcamp pages with custom domain
2188         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
2189         if mobj is not None:
2190             burl = unescapeHTML(mobj.group(1))
2191             # Don't set the extractor because it can be a track url or an album
2192             return self.url_result(burl)
2193
2194         # Look for embedded Vevo player
2195         mobj = re.search(
2196             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
2197         if mobj is not None:
2198             return self.url_result(mobj.group('url'))
2199
2200         # Look for embedded Viddler player
2201         mobj = re.search(
2202             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
2203             webpage)
2204         if mobj is not None:
2205             return self.url_result(mobj.group('url'))
2206
2207         # Look for NYTimes player
2208         mobj = re.search(
2209             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
2210             webpage)
2211         if mobj is not None:
2212             return self.url_result(mobj.group('url'))
2213
2214         # Look for Libsyn player
2215         mobj = re.search(
2216             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
2217         if mobj is not None:
2218             return self.url_result(mobj.group('url'))
2219
2220         # Look for Ooyala videos
2221         mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
2222                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
2223                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
2224                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
2225         if mobj is not None:
2226             embed_token = self._search_regex(
2227                 r'embedToken[\'"]?\s*:\s*[\'"]([^\'"]+)',
2228                 webpage, 'ooyala embed token', default=None)
2229             return OoyalaIE._build_url_result(smuggle_url(
2230                 mobj.group('ec'), {
2231                     'domain': url,
2232                     'embed_token': embed_token,
2233                 }))
2234
2235         # Look for multiple Ooyala embeds on SBN network websites
2236         mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
2237         if mobj is not None:
2238             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
2239             if embeds:
2240                 return self.playlist_from_matches(
2241                     embeds, video_id, video_title,
2242                     getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
2243
2244         # Look for Aparat videos
2245         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
2246         if mobj is not None:
2247             return self.url_result(mobj.group(1), 'Aparat')
2248
2249         # Look for MPORA videos
2250         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
2251         if mobj is not None:
2252             return self.url_result(mobj.group(1), 'Mpora')
2253
2254         # Look for embedded NovaMov-based player
2255         mobj = re.search(
2256             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
2257                     (?P<url>http://(?:(?:embed|www)\.)?
2258                         (?:novamov\.com|
2259                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
2260                            videoweed\.(?:es|com)|
2261                            movshare\.(?:net|sx|ag)|
2262                            divxstage\.(?:eu|net|ch|co|at|ag))
2263                         /embed\.php.+?)\1''', webpage)
2264         if mobj is not None:
2265             return self.url_result(mobj.group('url'))
2266
2267         # Look for embedded Facebook player
2268         facebook_urls = FacebookIE._extract_urls(webpage)
2269         if facebook_urls:
2270             return self.playlist_from_matches(facebook_urls, video_id, video_title)
2271
2272         # Look for embedded VK player
2273         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
2274         if mobj is not None:
2275             return self.url_result(mobj.group('url'), 'VK')
2276
2277         # Look for embedded Odnoklassniki player
2278         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1', webpage)
2279         if mobj is not None:
2280             return self.url_result(mobj.group('url'), 'Odnoklassniki')
2281
2282         # Look for embedded ivi player
2283         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
2284         if mobj is not None:
2285             return self.url_result(mobj.group('url'), 'Ivi')
2286
2287         # Look for embedded Huffington Post player
2288         mobj = re.search(
2289             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
2290         if mobj is not None:
2291             return self.url_result(mobj.group('url'), 'HuffPost')
2292
2293         # Look for embed.ly
2294         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
2295         if mobj is not None:
2296             return self.url_result(mobj.group('url'))
2297         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
2298         if mobj is not None:
2299             return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
2300
2301         # Look for funnyordie embed
2302         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
2303         if matches:
2304             return self.playlist_from_matches(
2305                 matches, video_id, video_title, getter=unescapeHTML, ie='FunnyOrDie')
2306
2307         # Look for BBC iPlayer embed
2308         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
2309         if matches:
2310             return self.playlist_from_matches(matches, video_id, video_title, ie='BBCCoUk')
2311
2312         # Look for embedded RUTV player
2313         rutv_url = RUTVIE._extract_url(webpage)
2314         if rutv_url:
2315             return self.url_result(rutv_url, 'RUTV')
2316
2317         # Look for embedded TVC player
2318         tvc_url = TVCIE._extract_url(webpage)
2319         if tvc_url:
2320             return self.url_result(tvc_url, 'TVC')
2321
2322         # Look for embedded SportBox player
2323         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
2324         if sportbox_urls:
2325             return self.playlist_from_matches(sportbox_urls, video_id, video_title, ie='SportBoxEmbed')
2326
2327         # Look for embedded XHamster player
2328         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
2329         if xhamster_urls:
2330             return self.playlist_from_matches(xhamster_urls, video_id, video_title, ie='XHamsterEmbed')
2331
2332         # Look for embedded TNAFlixNetwork player
2333         tnaflix_urls = TNAFlixNetworkEmbedIE._extract_urls(webpage)
2334         if tnaflix_urls:
2335             return self.playlist_from_matches(tnaflix_urls, video_id, video_title, ie=TNAFlixNetworkEmbedIE.ie_key())
2336
2337         # Look for embedded PornHub player
2338         pornhub_urls = PornHubIE._extract_urls(webpage)
2339         if pornhub_urls:
2340             return self.playlist_from_matches(pornhub_urls, video_id, video_title, ie=PornHubIE.ie_key())
2341
2342         # Look for embedded DrTuber player
2343         drtuber_urls = DrTuberIE._extract_urls(webpage)
2344         if drtuber_urls:
2345             return self.playlist_from_matches(drtuber_urls, video_id, video_title, ie=DrTuberIE.ie_key())
2346
2347         # Look for embedded RedTube player
2348         redtube_urls = RedTubeIE._extract_urls(webpage)
2349         if redtube_urls:
2350             return self.playlist_from_matches(redtube_urls, video_id, video_title, ie=RedTubeIE.ie_key())
2351
2352         # Look for embedded Tvigle player
2353         mobj = re.search(
2354             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
2355         if mobj is not None:
2356             return self.url_result(mobj.group('url'), 'Tvigle')
2357
2358         # Look for embedded TED player
2359         mobj = re.search(
2360             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
2361         if mobj is not None:
2362             return self.url_result(mobj.group('url'), 'TED')
2363
2364         # Look for embedded Ustream videos
2365         ustream_url = UstreamIE._extract_url(webpage)
2366         if ustream_url:
2367             return self.url_result(ustream_url, UstreamIE.ie_key())
2368
2369         # Look for embedded arte.tv player
2370         mobj = re.search(
2371             r'<(?:script|iframe) [^>]*?src="(?P<url>http://www\.arte\.tv/(?:playerv2/embed|arte_vp/index)[^"]+)"',
2372             webpage)
2373         if mobj is not None:
2374             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
2375
2376         # Look for embedded francetv player
2377         mobj = re.search(
2378             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
2379             webpage)
2380         if mobj is not None:
2381             return self.url_result(mobj.group('url'))
2382
2383         # Look for embedded smotri.com player
2384         smotri_url = SmotriIE._extract_url(webpage)
2385         if smotri_url:
2386             return self.url_result(smotri_url, 'Smotri')
2387
2388         # Look for embedded Myvi.ru player
2389         myvi_url = MyviIE._extract_url(webpage)
2390         if myvi_url:
2391             return self.url_result(myvi_url)
2392
2393         # Look for embedded soundcloud player
2394         soundcloud_urls = SoundcloudIE._extract_urls(webpage)
2395         if soundcloud_urls:
2396             return self.playlist_from_matches(soundcloud_urls, video_id, video_title, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
2397
2398         # Look for tunein player
2399         tunein_urls = TuneInBaseIE._extract_urls(webpage)
2400         if tunein_urls:
2401             return self.playlist_from_matches(tunein_urls, video_id, video_title)
2402
2403         # Look for embedded mtvservices player
2404         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
2405         if mtvservices_url:
2406             return self.url_result(mtvservices_url, ie='MTVServicesEmbedded')
2407
2408         # Look for embedded yahoo player
2409         mobj = re.search(
2410             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
2411             webpage)
2412         if mobj is not None:
2413             return self.url_result(mobj.group('url'), 'Yahoo')
2414
2415         # Look for embedded sbs.com.au player
2416         mobj = re.search(
2417             r'''(?x)
2418             (?:
2419                 <meta\s+property="og:video"\s+content=|
2420                 <iframe[^>]+?src=
2421             )
2422             (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
2423             webpage)
2424         if mobj is not None:
2425             return self.url_result(mobj.group('url'), 'SBS')
2426
2427         # Look for embedded Cinchcast player
2428         mobj = re.search(
2429             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
2430             webpage)
2431         if mobj is not None:
2432             return self.url_result(mobj.group('url'), 'Cinchcast')
2433
2434         mobj = re.search(
2435             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
2436             webpage)
2437         if not mobj:
2438             mobj = re.search(
2439                 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
2440                 webpage)
2441         if mobj is not None:
2442             return self.url_result(mobj.group('url'), 'MLB')
2443
2444         mobj = re.search(
2445             r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
2446             webpage)
2447         if mobj is not None:
2448             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
2449
2450         mobj = re.search(
2451             r'<iframe[^>]+src="(?P<url>https?://(?:new\.)?livestream\.com/[^"]+/player[^"]+)"',
2452             webpage)
2453         if mobj is not None:
2454             return self.url_result(mobj.group('url'), 'Livestream')
2455
2456         # Look for Zapiks embed
2457         mobj = re.search(
2458             r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
2459         if mobj is not None:
2460             return self.url_result(mobj.group('url'), 'Zapiks')
2461
2462         # Look for Kaltura embeds
2463         kaltura_url = KalturaIE._extract_url(webpage)
2464         if kaltura_url:
2465             return self.url_result(smuggle_url(kaltura_url, {'source_url': url}), KalturaIE.ie_key())
2466
2467         # Look for EaglePlatform embeds
2468         eagleplatform_url = EaglePlatformIE._extract_url(webpage)
2469         if eagleplatform_url:
2470             return self.url_result(smuggle_url(eagleplatform_url, {'referrer': url}), EaglePlatformIE.ie_key())
2471
2472         # Look for ClipYou (uses EaglePlatform) embeds
2473         mobj = re.search(
2474             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
2475         if mobj is not None:
2476             return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
2477
2478         # Look for Pladform embeds
2479         pladform_url = PladformIE._extract_url(webpage)
2480         if pladform_url:
2481             return self.url_result(pladform_url)
2482
2483         # Look for Videomore embeds
2484         videomore_url = VideomoreIE._extract_url(webpage)
2485         if videomore_url:
2486             return self.url_result(videomore_url)
2487
2488         # Look for Webcaster embeds
2489         webcaster_url = WebcasterFeedIE._extract_url(self, webpage)
2490         if webcaster_url:
2491             return self.url_result(webcaster_url, ie=WebcasterFeedIE.ie_key())
2492
2493         # Look for Playwire embeds
2494         mobj = re.search(
2495             r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
2496         if mobj is not None:
2497             return self.url_result(mobj.group('url'))
2498
2499         # Look for 5min embeds
2500         mobj = re.search(
2501             r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
2502         if mobj is not None:
2503             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
2504
2505         # Look for Crooks and Liars embeds
2506         mobj = re.search(
2507             r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
2508         if mobj is not None:
2509             return self.url_result(mobj.group('url'))
2510
2511         # Look for NBC Sports VPlayer embeds
2512         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
2513         if nbc_sports_url:
2514             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
2515
2516         # Look for NBC News embeds
2517         nbc_news_embed_url = re.search(
2518             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//www\.nbcnews\.com/widget/video-embed/[^"\']+)\1', webpage)
2519         if nbc_news_embed_url:
2520             return self.url_result(nbc_news_embed_url.group('url'), 'NBCNews')
2521
2522         # Look for Google Drive embeds
2523         google_drive_url = GoogleDriveIE._extract_url(webpage)
2524         if google_drive_url:
2525             return self.url_result(google_drive_url, 'GoogleDrive')
2526
2527         # Look for UDN embeds
2528         mobj = re.search(
2529             r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage)
2530         if mobj is not None:
2531             return self.url_result(
2532                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
2533
2534         # Look for Senate ISVP iframe
2535         senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
2536         if senate_isvp_url:
2537             return self.url_result(senate_isvp_url, 'SenateISVP')
2538
2539         # Look for Dailymotion Cloud videos
2540         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
2541         if dmcloud_url:
2542             return self.url_result(dmcloud_url, 'DailymotionCloud')
2543
2544         # Look for OnionStudios embeds
2545         onionstudios_url = OnionStudiosIE._extract_url(webpage)
2546         if onionstudios_url:
2547             return self.url_result(onionstudios_url)
2548
2549         # Look for ViewLift embeds
2550         viewlift_url = ViewLiftEmbedIE._extract_url(webpage)
2551         if viewlift_url:
2552             return self.url_result(viewlift_url)
2553
2554         # Look for JWPlatform embeds
2555         jwplatform_url = JWPlatformIE._extract_url(webpage)
2556         if jwplatform_url:
2557             return self.url_result(jwplatform_url, 'JWPlatform')
2558
2559         # Look for Digiteka embeds
2560         digiteka_url = DigitekaIE._extract_url(webpage)
2561         if digiteka_url:
2562             return self.url_result(self._proto_relative_url(digiteka_url), DigitekaIE.ie_key())
2563
2564         # Look for Arkena embeds
2565         arkena_url = ArkenaIE._extract_url(webpage)
2566         if arkena_url:
2567             return self.url_result(arkena_url, ArkenaIE.ie_key())
2568
2569         # Look for Piksel embeds
2570         piksel_url = PikselIE._extract_url(webpage)
2571         if piksel_url:
2572             return self.url_result(piksel_url, PikselIE.ie_key())
2573
2574         # Look for Limelight embeds
2575         limelight_urls = LimelightBaseIE._extract_urls(webpage, url)
2576         if limelight_urls:
2577             return self.playlist_result(
2578                 limelight_urls, video_id, video_title, video_description)
2579
2580         # Look for Anvato embeds
2581         anvato_urls = AnvatoIE._extract_urls(self, webpage, video_id)
2582         if anvato_urls:
2583             return self.playlist_result(
2584                 anvato_urls, video_id, video_title, video_description)
2585
2586         # Look for AdobeTVVideo embeds
2587         mobj = re.search(
2588             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
2589             webpage)
2590         if mobj is not None:
2591             return self.url_result(
2592                 self._proto_relative_url(unescapeHTML(mobj.group(1))),
2593                 'AdobeTVVideo')
2594
2595         # Look for Vine embeds
2596         mobj = re.search(
2597             r'<iframe[^>]+src=[\'"]((?:https?:)?//(?:www\.)?vine\.co/v/[^/]+/embed/(?:simple|postcard))',
2598             webpage)
2599         if mobj is not None:
2600             return self.url_result(
2601                 self._proto_relative_url(unescapeHTML(mobj.group(1))), 'Vine')
2602
2603         # Look for VODPlatform embeds
2604         mobj = re.search(
2605             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vod-platform\.net/[eE]mbed/.+?)\1',
2606             webpage)
2607         if mobj is not None:
2608             return self.url_result(
2609                 self._proto_relative_url(unescapeHTML(mobj.group('url'))), 'VODPlatform')
2610
2611         # Look for Mangomolo embeds
2612         mobj = re.search(
2613             r'''(?x)<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?admin\.mangomolo\.com/analytics/index\.php/customers/embed/
2614                 (?:
2615                     video\?.*?\bid=(?P<video_id>\d+)|
2616                     index\?.*?\bchannelid=(?P<channel_id>(?:[A-Za-z0-9+/=]|%2B|%2F|%3D)+)
2617                 ).+?)\1''', webpage)
2618         if mobj is not None:
2619             info = {
2620                 '_type': 'url_transparent',
2621                 'url': self._proto_relative_url(unescapeHTML(mobj.group('url'))),
2622                 'title': video_title,
2623                 'description': video_description,
2624                 'thumbnail': video_thumbnail,
2625                 'uploader': video_uploader,
2626             }
2627             video_id = mobj.group('video_id')
2628             if video_id:
2629                 info.update({
2630                     'ie_key': 'MangomoloVideo',
2631                     'id': video_id,
2632                 })
2633             else:
2634                 info.update({
2635                     'ie_key': 'MangomoloLive',
2636                     'id': mobj.group('channel_id'),
2637                 })
2638             return info
2639
2640         # Look for Instagram embeds
2641         instagram_embed_url = InstagramIE._extract_embed_url(webpage)
2642         if instagram_embed_url is not None:
2643             return self.url_result(
2644                 self._proto_relative_url(instagram_embed_url), InstagramIE.ie_key())
2645
2646         # Look for LiveLeak embeds
2647         liveleak_url = LiveLeakIE._extract_url(webpage)
2648         if liveleak_url:
2649             return self.url_result(liveleak_url, 'LiveLeak')
2650
2651         # Look for 3Q SDN embeds
2652         threeqsdn_url = ThreeQSDNIE._extract_url(webpage)
2653         if threeqsdn_url:
2654             return {
2655                 '_type': 'url_transparent',
2656                 'ie_key': ThreeQSDNIE.ie_key(),
2657                 'url': self._proto_relative_url(threeqsdn_url),
2658                 'title': video_title,
2659                 'description': video_description,
2660                 'thumbnail': video_thumbnail,
2661                 'uploader': video_uploader,
2662             }
2663
2664         # Look for VBOX7 embeds
2665         vbox7_url = Vbox7IE._extract_url(webpage)
2666         if vbox7_url:
2667             return self.url_result(vbox7_url, Vbox7IE.ie_key())
2668
2669         # Look for DBTV embeds
2670         dbtv_urls = DBTVIE._extract_urls(webpage)
2671         if dbtv_urls:
2672             return self.playlist_from_matches(dbtv_urls, video_id, video_title, ie=DBTVIE.ie_key())
2673
2674         # Look for Videa embeds
2675         videa_urls = VideaIE._extract_urls(webpage)
2676         if videa_urls:
2677             return self.playlist_from_matches(videa_urls, video_id, video_title, ie=VideaIE.ie_key())
2678
2679         # Look for 20 minuten embeds
2680         twentymin_urls = TwentyMinutenIE._extract_urls(webpage)
2681         if twentymin_urls:
2682             return self.playlist_from_matches(
2683                 twentymin_urls, video_id, video_title, ie=TwentyMinutenIE.ie_key())
2684
2685         # Look for Openload embeds
2686         openload_urls = OpenloadIE._extract_urls(webpage)
2687         if openload_urls:
2688             return self.playlist_from_matches(
2689                 openload_urls, video_id, video_title, ie=OpenloadIE.ie_key())
2690
2691         # Look for VideoPress embeds
2692         videopress_urls = VideoPressIE._extract_urls(webpage)
2693         if videopress_urls:
2694             return self.playlist_from_matches(
2695                 videopress_urls, video_id, video_title, ie=VideoPressIE.ie_key())
2696
2697         # Look for Rutube embeds
2698         rutube_urls = RutubeIE._extract_urls(webpage)
2699         if rutube_urls:
2700             return self.playlist_from_matches(
2701                 rutube_urls, ie=RutubeIE.ie_key())
2702
2703         # Look for WashingtonPost embeds
2704         wapo_urls = WashingtonPostIE._extract_urls(webpage)
2705         if wapo_urls:
2706             return self.playlist_from_matches(
2707                 wapo_urls, video_id, video_title, ie=WashingtonPostIE.ie_key())
2708
2709         # Look for Mediaset embeds
2710         mediaset_urls = MediasetIE._extract_urls(webpage)
2711         if mediaset_urls:
2712             return self.playlist_from_matches(
2713                 mediaset_urls, video_id, video_title, ie=MediasetIE.ie_key())
2714
2715         def merge_dicts(dict1, dict2):
2716             merged = {}
2717             for k, v in dict1.items():
2718                 if v is not None:
2719                     merged[k] = v
2720             for k, v in dict2.items():
2721                 if v is None:
2722                     continue
2723                 if (k not in merged or
2724                         (isinstance(v, compat_str) and v and
2725                             isinstance(merged[k], compat_str) and
2726                             not merged[k])):
2727                     merged[k] = v
2728             return merged
2729
2730         # Looking for http://schema.org/VideoObject
2731         json_ld = self._search_json_ld(
2732             webpage, video_id, default={}, expected_type='VideoObject')
2733         if json_ld.get('url'):
2734             return merge_dicts(json_ld, info_dict)
2735
2736         # Look for HTML5 media
2737         entries = self._parse_html5_media_entries(url, webpage, video_id, m3u8_id='hls')
2738         if entries:
2739             for entry in entries:
2740                 entry.update({
2741                     'id': video_id,
2742                     'title': video_title,
2743                 })
2744                 self._sort_formats(entry['formats'])
2745             return self.playlist_result(entries)
2746
2747         jwplayer_data = self._find_jwplayer_data(
2748             webpage, video_id, transform_source=js_to_json)
2749         if jwplayer_data:
2750             info = self._parse_jwplayer_data(
2751                 jwplayer_data, video_id, require_title=False, base_url=url)
2752             return merge_dicts(info, info_dict)
2753
2754         def check_video(vurl):
2755             if YoutubeIE.suitable(vurl):
2756                 return True
2757             if RtmpIE.suitable(vurl):
2758                 return True
2759             vpath = compat_urlparse.urlparse(vurl).path
2760             vext = determine_ext(vpath)
2761             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml', 'js', 'xml')
2762
2763         def filter_video(urls):
2764             return list(filter(check_video, urls))
2765
2766         # Start with something easy: JW Player in SWFObject
2767         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
2768         if not found:
2769             # Look for gorilla-vid style embedding
2770             found = filter_video(re.findall(r'''(?sx)
2771                 (?:
2772                     jw_plugins|
2773                     JWPlayerOptions|
2774                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
2775                 )
2776                 .*?
2777                 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
2778         if not found:
2779             # Broaden the search a little bit
2780             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
2781         if not found:
2782             # Broaden the findall a little bit: JWPlayer JS loader
2783             found = filter_video(re.findall(
2784                 r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
2785         if not found:
2786             # Flow player
2787             found = filter_video(re.findall(r'''(?xs)
2788                 flowplayer\("[^"]+",\s*
2789                     \{[^}]+?\}\s*,
2790                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
2791                         ["']?url["']?\s*:\s*["']([^"']+)["']
2792             ''', webpage))
2793         if not found:
2794             # Cinerama player
2795             found = re.findall(
2796                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
2797         if not found:
2798             # Try to find twitter cards info
2799             # twitter:player:stream should be checked before twitter:player since
2800             # it is expected to contain a raw stream (see
2801             # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser)
2802             found = filter_video(re.findall(
2803                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
2804         if not found:
2805             # We look for Open Graph info:
2806             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
2807             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
2808             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
2809             if m_video_type is not None:
2810                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
2811         if not found:
2812             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
2813             found = re.search(
2814                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
2815                 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
2816                 webpage)
2817             if not found:
2818                 # Look also in Refresh HTTP header
2819                 refresh_header = head_response.headers.get('Refresh')
2820                 if refresh_header:
2821                     # In python 2 response HTTP headers are bytestrings
2822                     if sys.version_info < (3, 0) and isinstance(refresh_header, str):
2823                         refresh_header = refresh_header.decode('iso-8859-1')
2824                     found = re.search(REDIRECT_REGEX, refresh_header)
2825             if found:
2826                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
2827                 if new_url != url:
2828                     self.report_following_redirect(new_url)
2829                     return {
2830                         '_type': 'url',
2831                         'url': new_url,
2832                     }
2833                 else:
2834                     found = None
2835
2836         if not found:
2837             # twitter:player is a https URL to iframe player that may or may not
2838             # be supported by youtube-dl thus this is checked the very last (see
2839             # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser)
2840             embed_url = self._html_search_meta('twitter:player', webpage, default=None)
2841             if embed_url:
2842                 return self.url_result(embed_url)
2843
2844         if not found:
2845             raise UnsupportedError(url)
2846
2847         entries = []
2848         for video_url in orderedSet(found):
2849             video_url = unescapeHTML(video_url)
2850             video_url = video_url.replace('\\/', '/')
2851             video_url = compat_urlparse.urljoin(url, video_url)
2852             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
2853
2854             # Sometimes, jwplayer extraction will result in a YouTube URL
2855             if YoutubeIE.suitable(video_url):
2856                 entries.append(self.url_result(video_url, 'Youtube'))
2857                 continue
2858
2859             # here's a fun little line of code for you:
2860             video_id = os.path.splitext(video_id)[0]
2861
2862             entry_info_dict = {
2863                 'id': video_id,
2864                 'uploader': video_uploader,
2865                 'title': video_title,
2866                 'age_limit': age_limit,
2867             }
2868
2869             if RtmpIE.suitable(video_url):
2870                 entry_info_dict.update({
2871                     '_type': 'url_transparent',
2872                     'ie_key': RtmpIE.ie_key(),
2873                     'url': video_url,
2874                 })
2875                 entries.append(entry_info_dict)
2876                 continue
2877
2878             ext = determine_ext(video_url)
2879             if ext == 'smil':
2880                 entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
2881             elif ext == 'xspf':
2882                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
2883             elif ext == 'm3u8':
2884                 entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
2885             elif ext == 'mpd':
2886                 entry_info_dict['formats'] = self._extract_mpd_formats(video_url, video_id)
2887             elif ext == 'f4m':
2888                 entry_info_dict['formats'] = self._extract_f4m_formats(video_url, video_id)
2889             elif re.search(r'(?i)\.(?:ism|smil)/manifest', video_url) and video_url != url:
2890                 # Just matching .ism/manifest is not enough to be reliably sure
2891                 # whether it's actually an ISM manifest or some other streaming
2892                 # manifest since there are various streaming URL formats
2893                 # possible (see [1]) as well as some other shenanigans like
2894                 # .smil/manifest URLs that actually serve an ISM (see [2]) and
2895                 # so on.
2896                 # Thus the most reasonable way to solve this is to delegate
2897                 # to generic extractor in order to look into the contents of
2898                 # the manifest itself.
2899                 # 1. https://azure.microsoft.com/en-us/documentation/articles/media-services-deliver-content-overview/#streaming-url-formats
2900                 # 2. https://svs.itworkscdn.net/lbcivod/smil:itwfcdn/lbci/170976.smil/Manifest
2901                 entry_info_dict = self.url_result(
2902                     smuggle_url(video_url, {'to_generic': True}),
2903                     GenericIE.ie_key())
2904             else:
2905                 entry_info_dict['url'] = video_url
2906
2907             if entry_info_dict.get('formats'):
2908                 self._sort_formats(entry_info_dict['formats'])
2909
2910             entries.append(entry_info_dict)
2911
2912         if len(entries) == 1:
2913             return entries[0]
2914         else:
2915             for num, e in enumerate(entries, start=1):
2916                 # 'url' results don't have a title
2917                 if e.get('title') is not None:
2918                     e['title'] = '%s (%d)' % (e['title'], num)
2919             return {
2920                 '_type': 'playlist',
2921                 'entries': entries,
2922             }