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

   1 # encoding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 import os
   6 import re
   7
   8 from .common import InfoExtractor
   9 from .youtube import YoutubeIE
  10 from ..compat import (
  11     compat_urllib_parse,
  12     compat_urlparse,
  13     compat_xml_parse_error,
  14 )
  15 from ..utils import (
  16     determine_ext,
  17     ExtractorError,
  18     float_or_none,
  19     HEADRequest,
  20     is_html,
  21     orderedSet,
  22     parse_xml,
  23     smuggle_url,
  24     unescapeHTML,
  25     unified_strdate,
  26     unsmuggle_url,
  27     UnsupportedError,
  28     url_basename,
  29 )
  30 from .brightcove import BrightcoveIE
  31 from .ooyala import OoyalaIE
  32 from .rutv import RUTVIE
  33 from .smotri import SmotriIE
  34 from .condenast import CondeNastIE
  35
  36
  37 class GenericIE(InfoExtractor):
  38     IE_DESC = 'Generic downloader that works on some sites'
  39     _VALID_URL = r'.*'
  40     IE_NAME = 'generic'
  41     _TESTS = [
  42         {
  43             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
  44             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
  45             'info_dict': {
  46                 'id': '13601338388002',
  47                 'ext': 'mp4',
  48                 'uploader': 'www.hodiho.fr',
  49                 'title': 'R\u00e9gis plante sa Jeep',
  50             }
  51         },
  52         # bandcamp page with custom domain
  53         {
  54             'add_ie': ['Bandcamp'],
  55             'url': 'http://bronyrock.com/track/the-pony-mash',
  56             'info_dict': {
  57                 'id': '3235767654',
  58                 'ext': 'mp3',
  59                 'title': 'The Pony Mash',
  60                 'uploader': 'M_Pallante',
  61             },
  62             'skip': 'There is a limit of 200 free downloads / month for the test song',
  63         },
  64         # embedded brightcove video
  65         # it also tests brightcove videos that need to set the 'Referer' in the
  66         # http requests
  67         {
  68             'add_ie': ['Brightcove'],
  69             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
  70             'info_dict': {
  71                 'id': '2765128793001',
  72                 'ext': 'mp4',
  73                 'title': 'Le cours de bourse : l’analyse technique',
  74                 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
  75                 'uploader': 'BFM BUSINESS',
  76             },
  77             'params': {
  78                 'skip_download': True,
  79             },
  80         },
  81         {
  82             # https://github.com/rg3/youtube-dl/issues/2253
  83             'url': 'http://bcove.me/i6nfkrc3',
  84             'md5': '0ba9446db037002366bab3b3eb30c88c',
  85             'info_dict': {
  86                 'id': '3101154703001',
  87                 'ext': 'mp4',
  88                 'title': 'Still no power',
  89                 'uploader': 'thestar.com',
  90                 '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.',
  91             },
  92             'add_ie': ['Brightcove'],
  93         },
  94         {
  95             'url': 'http://www.championat.com/video/football/v/87/87499.html',
  96             'md5': 'fb973ecf6e4a78a67453647444222983',
  97             'info_dict': {
  98                 'id': '3414141473001',
  99                 'ext': 'mp4',
 100                 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
 101                 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
 102                 'uploader': 'Championat',
 103             },
 104         },
 105         {
 106             # https://github.com/rg3/youtube-dl/issues/3541
 107             'add_ie': ['Brightcove'],
 108             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
 109             'info_dict': {
 110                 'id': '3866516442001',
 111                 'ext': 'mp4',
 112                 'title': 'Leer mij vrouwen kennen: Aflevering 1',
 113                 'description': 'Leer mij vrouwen kennen: Aflevering 1',
 114                 'uploader': 'SBS Broadcasting',
 115             },
 116             'skip': 'Restricted to Netherlands',
 117             'params': {
 118                 'skip_download': True,  # m3u8 download
 119             },
 120         },
 121         # Direct link to a video
 122         {
 123             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
 124             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
 125             'info_dict': {
 126                 'id': 'trailer',
 127                 'ext': 'mp4',
 128                 'title': 'trailer',
 129                 'upload_date': '20100513',
 130             }
 131         },
 132         # ooyala video
 133         {
 134             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
 135             'md5': '166dd577b433b4d4ebfee10b0824d8ff',
 136             'info_dict': {
 137                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
 138                 'ext': 'mp4',
 139                 'title': '2cc213299525360.mov',  # that's what we get
 140             },
 141             'add_ie': ['Ooyala'],
 142         },
 143         # google redirect
 144         {
 145             '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',
 146             'info_dict': {
 147                 'id': 'cmQHVoWB5FY',
 148                 'ext': 'mp4',
 149                 'upload_date': '20130224',
 150                 'uploader_id': 'TheVerge',
 151                 'description': 're:^Chris Ziegler takes a look at the\.*',
 152                 'uploader': 'The Verge',
 153                 'title': 'First Firefox OS phones side-by-side',
 154             },
 155             'params': {
 156                 'skip_download': False,
 157             }
 158         },
 159         # embed.ly video
 160         {
 161             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
 162             'info_dict': {
 163                 'id': '9ODmcdjQcHQ',
 164                 'ext': 'mp4',
 165                 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
 166                 'upload_date': '20140225',
 167                 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
 168                 'uploader': 'Tested',
 169                 'uploader_id': 'testedcom',
 170             },
 171             # No need to test YoutubeIE here
 172             'params': {
 173                 'skip_download': True,
 174             },
 175         },
 176         # funnyordie embed
 177         {
 178             'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
 179             'info_dict': {
 180                 'id': '18e820ec3f',
 181                 'ext': 'mp4',
 182                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
 183                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
 184             },
 185         },
 186         # BBC iPlayer embeds
 187         {
 188             'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
 189             'info_dict': {
 190                 'title': 'BBC - Blogs -  Adam Curtis - BUGGER',
 191             },
 192             'playlist_mincount': 18,
 193         },
 194         # RUTV embed
 195         {
 196             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
 197             'info_dict': {
 198                 'id': '776940',
 199                 'ext': 'mp4',
 200                 'title': 'Охотское море стало целиком российским',
 201                 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
 202             },
 203             'params': {
 204                 # m3u8 download
 205                 'skip_download': True,
 206             },
 207         },
 208         # Embedded TED video
 209         {
 210             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
 211             'md5': '65fdff94098e4a607385a60c5177c638',
 212             'info_dict': {
 213                 'id': '1969',
 214                 'ext': 'mp4',
 215                 'title': 'Hidden miracles of the natural world',
 216                 'uploader': 'Louie Schwartzberg',
 217                 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
 218             }
 219         },
 220         # Embeded Ustream video
 221         {
 222             'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
 223             'md5': '27b99cdb639c9b12a79bca876a073417',
 224             'info_dict': {
 225                 'id': '45734260',
 226                 'ext': 'flv',
 227                 'uploader': 'AU SPA:  The NSA and Privacy',
 228                 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
 229             }
 230         },
 231         # nowvideo embed hidden behind percent encoding
 232         {
 233             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
 234             'md5': '2baf4ddd70f697d94b1c18cf796d5107',
 235             'info_dict': {
 236                 'id': '06e53103ca9aa',
 237                 'ext': 'flv',
 238                 'title': 'Macross Episode 001  Watch Macross Episode 001 onl',
 239                 'description': 'No description',
 240             },
 241         },
 242         # arte embed
 243         {
 244             'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
 245             'md5': '7653032cbb25bf6c80d80f217055fa43',
 246             'info_dict': {
 247                 'id': '048195-004_PLUS7-F',
 248                 'ext': 'flv',
 249                 'title': 'X:enius',
 250                 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
 251                 'upload_date': '20140320',
 252             },
 253             'params': {
 254                 'skip_download': 'Requires rtmpdump'
 255             }
 256         },
 257         # Condé Nast embed
 258         {
 259             'url': 'http://www.wired.com/2014/04/honda-asimo/',
 260             'md5': 'ba0dfe966fa007657bd1443ee672db0f',
 261             'info_dict': {
 262                 'id': '53501be369702d3275860000',
 263                 'ext': 'mp4',
 264                 'title': 'Honda’s  New Asimo Robot Is More Human Than Ever',
 265             }
 266         },
 267         # Dailymotion embed
 268         {
 269             'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
 270             'md5': '441aeeb82eb72c422c7f14ec533999cd',
 271             'info_dict': {
 272                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
 273                 'ext': 'mp4',
 274                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
 275                 'uploader': 'Spi0n',
 276             },
 277             'add_ie': ['Dailymotion'],
 278         },
 279         # YouTube embed
 280         {
 281             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
 282             'info_dict': {
 283                 'id': 'FXRb4ykk4S0',
 284                 'ext': 'mp4',
 285                 'title': 'The NBL Auction 2014',
 286                 'uploader': 'BADMINTON England',
 287                 'uploader_id': 'BADMINTONEvents',
 288                 'upload_date': '20140603',
 289                 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
 290             },
 291             'add_ie': ['Youtube'],
 292             'params': {
 293                 'skip_download': True,
 294             }
 295         },
 296         # MTVSercices embed
 297         {
 298             'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
 299             'md5': '35727f82f58c76d996fc188f9755b0d5',
 300             'info_dict': {
 301                 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
 302                 'ext': 'mp4',
 303                 'title': 'Review',
 304                 'description': 'Mario\'s life in the fast lane has never looked so good.',
 305             },
 306         },
 307         # YouTube embed via <data-embed-url="">
 308         {
 309             'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
 310             'info_dict': {
 311                 'id': '4vAffPZIT44',
 312                 'ext': 'mp4',
 313                 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
 314                 'uploader': 'Gameloft',
 315                 'uploader_id': 'gameloft',
 316                 'upload_date': '20140828',
 317                 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
 318             },
 319             'params': {
 320                 'skip_download': True,
 321             }
 322         },
 323         # Camtasia studio
 324         {
 325             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
 326             'playlist': [{
 327                 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
 328                 'info_dict': {
 329                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 330                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
 331                     'ext': 'flv',
 332                     'duration': 2235.90,
 333                 }
 334             }, {
 335                 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
 336                 'info_dict': {
 337                     'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
 338                     'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
 339                     'ext': 'flv',
 340                     'duration': 2235.93,
 341                 }
 342             }],
 343             'info_dict': {
 344                 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
 345             }
 346         },
 347         # Flowplayer
 348         {
 349             'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
 350             'md5': '9d65602bf31c6e20014319c7d07fba27',
 351             'info_dict': {
 352                 'id': '5123ea6d5e5a7',
 353                 'ext': 'mp4',
 354                 'age_limit': 18,
 355                 'uploader': 'www.handjobhub.com',
 356                 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
 357             }
 358         },
 359         # RSS feed
 360         {
 361             'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 362             'info_dict': {
 363                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
 364                 'title': 'Zero Punctuation',
 365                 'description': 're:.*groundbreaking video review series.*'
 366             },
 367             'playlist_mincount': 11,
 368         },
 369         # Multiple brightcove videos
 370         # https://github.com/rg3/youtube-dl/issues/2283
 371         {
 372             'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
 373             'info_dict': {
 374                 'id': 'always-never',
 375                 'title': 'Always / Never - The New Yorker',
 376             },
 377             'playlist_count': 3,
 378             'params': {
 379                 'extract_flat': False,
 380                 'skip_download': True,
 381             }
 382         },
 383         # MLB embed
 384         {
 385             'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
 386             'md5': '96f09a37e44da40dd083e12d9a683327',
 387             'info_dict': {
 388                 'id': '33322633',
 389                 'ext': 'mp4',
 390                 'title': 'Ump changes call to ball',
 391                 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
 392                 'duration': 48,
 393                 'timestamp': 1401537900,
 394                 'upload_date': '20140531',
 395                 'thumbnail': 're:^https?://.*\.jpg$',
 396             },
 397         },
 398         # Wistia embed
 399         {
 400             'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 401             'md5': '8788b683c777a5cf25621eaf286d0c23',
 402             'info_dict': {
 403                 'id': '1cfaf6b7ea',
 404                 'ext': 'mov',
 405                 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
 406                 'duration': 643.0,
 407                 'filesize': 182808282,
 408                 'uploader': 'education-portal.com',
 409             },
 410         },
 411         {
 412             'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
 413             'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
 414             'info_dict': {
 415                 'id': 'uxjb0lwrcz',
 416                 'ext': 'mp4',
 417                 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
 418                 'duration': 1715.0,
 419                 'uploader': 'thoughtworks.wistia.com',
 420             },
 421         },
 422         # Direct download with broken HEAD
 423         {
 424             'url': 'http://ai-radio.org:8000/radio.opus',
 425             'info_dict': {
 426                 'id': 'radio',
 427                 'ext': 'opus',
 428                 'title': 'radio',
 429             },
 430             'params': {
 431                 'skip_download': True,  # infinite live stream
 432             },
 433             'expected_warnings': [
 434                 r'501.*Not Implemented'
 435             ],
 436         },
 437         # Soundcloud embed
 438         {
 439             'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
 440             'info_dict': {
 441                 'id': '174391317',
 442                 'ext': 'mp3',
 443                 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
 444                 'uploader': 'Sophos Security',
 445                 'title': 'Chet Chat 171 - Oct 29, 2014',
 446                 'upload_date': '20141029',
 447             }
 448         },
 449         # Livestream embed
 450         {
 451             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
 452             'info_dict': {
 453                 'id': '67864563',
 454                 'ext': 'flv',
 455                 'upload_date': '20141112',
 456                 'title': 'Rosetta #CometLanding webcast HL 10',
 457             }
 458         },
 459         # LazyYT
 460         {
 461             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
 462             'info_dict': {
 463                 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
 464             },
 465             'playlist_mincount': 2,
 466         },
 467         # Direct link with incorrect MIME type
 468         {
 469             'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 470             'md5': '4ccbebe5f36706d85221f204d7eb5913',
 471             'info_dict': {
 472                 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
 473                 'id': '5_Lennart_Poettering_-_Systemd',
 474                 'ext': 'webm',
 475                 'title': '5_Lennart_Poettering_-_Systemd',
 476                 'upload_date': '20141120',
 477             },
 478             'expected_warnings': [
 479                 'URL could be a direct video link, returning it as such.'
 480             ]
 481         },
 482         # Cinchcast embed
 483         {
 484             'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
 485             'info_dict': {
 486                 'id': '7141703',
 487                 'ext': 'mp3',
 488                 'upload_date': '20141126',
 489                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
 490             }
 491         },
 492         # Cinerama player
 493         {
 494             'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
 495             'info_dict': {
 496                 'id': '730m_DandD_1901_512k',
 497                 'ext': 'mp4',
 498                 'uploader': 'www.abc.net.au',
 499                 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
 500             }
 501         },
 502         # embedded viddler video
 503         {
 504             'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
 505             'info_dict': {
 506                 'id': '4d03aad9',
 507                 'ext': 'mp4',
 508                 'uploader': 'deadspin',
 509                 'title': 'WALL-TO-GORTAT',
 510                 'timestamp': 1422285291,
 511                 'upload_date': '20150126',
 512             },
 513             'add_ie': ['Viddler'],
 514         }
 515     ]
 516
 517     def report_following_redirect(self, new_url):
 518         """Report information extraction."""
 519         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
 520
 521     def _extract_rss(self, url, video_id, doc):
 522         playlist_title = doc.find('./channel/title').text
 523         playlist_desc_el = doc.find('./channel/description')
 524         playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
 525
 526         entries = [{
 527             '_type': 'url',
 528             'url': e.find('link').text,
 529             'title': e.find('title').text,
 530         } for e in doc.findall('./channel/item')]
 531
 532         return {
 533             '_type': 'playlist',
 534             'id': url,
 535             'title': playlist_title,
 536             'description': playlist_desc,
 537             'entries': entries,
 538         }
 539
 540     def _extract_camtasia(self, url, video_id, webpage):
 541         """ Returns None if no camtasia video can be found. """
 542
 543         camtasia_cfg = self._search_regex(
 544             r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
 545             webpage, 'camtasia configuration file', default=None)
 546         if camtasia_cfg is None:
 547             return None
 548
 549         title = self._html_search_meta('DC.title', webpage, fatal=True)
 550
 551         camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
 552         camtasia_cfg = self._download_xml(
 553             camtasia_url, video_id,
 554             note='Downloading camtasia configuration',
 555             errnote='Failed to download camtasia configuration')
 556         fileset_node = camtasia_cfg.find('./playlist/array/fileset')
 557
 558         entries = []
 559         for n in fileset_node.getchildren():
 560             url_n = n.find('./uri')
 561             if url_n is None:
 562                 continue
 563
 564             entries.append({
 565                 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
 566                 'title': '%s - %s' % (title, n.tag),
 567                 'url': compat_urlparse.urljoin(url, url_n.text),
 568                 'duration': float_or_none(n.find('./duration').text),
 569             })
 570
 571         return {
 572             '_type': 'playlist',
 573             'entries': entries,
 574             'title': title,
 575         }
 576
 577     def _real_extract(self, url):
 578         if url.startswith('//'):
 579             return {
 580                 '_type': 'url',
 581                 'url': self.http_scheme() + url,
 582             }
 583
 584         parsed_url = compat_urlparse.urlparse(url)
 585         if not parsed_url.scheme:
 586             default_search = self._downloader.params.get('default_search')
 587             if default_search is None:
 588                 default_search = 'fixup_error'
 589
 590             if default_search in ('auto', 'auto_warning', 'fixup_error'):
 591                 if '/' in url:
 592                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
 593                     return self.url_result('http://' + url)
 594                 elif default_search != 'fixup_error':
 595                     if default_search == 'auto_warning':
 596                         if re.match(r'^(?:url|URL)$', url):
 597                             raise ExtractorError(
 598                                 'Invalid URL:  %r . Call youtube-dl like this:  youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc"  ' % url,
 599                                 expected=True)
 600                         else:
 601                             self._downloader.report_warning(
 602                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
 603                     return self.url_result('ytsearch:' + url)
 604
 605             if default_search in ('error', 'fixup_error'):
 606                 raise ExtractorError(
 607                     '%r is not a valid URL. '
 608                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
 609                     % (url, url), expected=True)
 610             else:
 611                 if ':' not in default_search:
 612                     default_search += ':'
 613                 return self.url_result(default_search + url)
 614
 615         url, smuggled_data = unsmuggle_url(url)
 616         force_videoid = None
 617         is_intentional = smuggled_data and smuggled_data.get('to_generic')
 618         if smuggled_data and 'force_videoid' in smuggled_data:
 619             force_videoid = smuggled_data['force_videoid']
 620             video_id = force_videoid
 621         else:
 622             video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
 623
 624         self.to_screen('%s: Requesting header' % video_id)
 625
 626         head_req = HEADRequest(url)
 627         head_response = self._request_webpage(
 628             head_req, video_id,
 629             note=False, errnote='Could not send HEAD request to %s' % url,
 630             fatal=False)
 631
 632         if head_response is not False:
 633             # Check for redirect
 634             new_url = head_response.geturl()
 635             if url != new_url:
 636                 self.report_following_redirect(new_url)
 637                 if force_videoid:
 638                     new_url = smuggle_url(
 639                         new_url, {'force_videoid': force_videoid})
 640                 return self.url_result(new_url)
 641
 642         full_response = None
 643         if head_response is False:
 644             full_response = self._request_webpage(url, video_id)
 645             head_response = full_response
 646
 647         # Check for direct link to a video
 648         content_type = head_response.headers.get('Content-Type', '')
 649         m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
 650         if m:
 651             upload_date = unified_strdate(
 652                 head_response.headers.get('Last-Modified'))
 653             return {
 654                 'id': video_id,
 655                 'title': os.path.splitext(url_basename(url))[0],
 656                 'direct': True,
 657                 'formats': [{
 658                     'format_id': m.group('format_id'),
 659                     'url': url,
 660                     'vcodec': 'none' if m.group('type') == 'audio' else None
 661                 }],
 662                 'upload_date': upload_date,
 663             }
 664
 665         if not self._downloader.params.get('test', False) and not is_intentional:
 666             self._downloader.report_warning('Falling back on generic information extractor.')
 667
 668         if not full_response:
 669             full_response = self._request_webpage(url, video_id)
 670
 671         # Maybe it's a direct link to a video?
 672         # Be careful not to download the whole thing!
 673         first_bytes = full_response.read(512)
 674         if not is_html(first_bytes):
 675             self._downloader.report_warning(
 676                 'URL could be a direct video link, returning it as such.')
 677             upload_date = unified_strdate(
 678                 head_response.headers.get('Last-Modified'))
 679             return {
 680                 'id': video_id,
 681                 'title': os.path.splitext(url_basename(url))[0],
 682                 'direct': True,
 683                 'url': url,
 684                 'upload_date': upload_date,
 685             }
 686
 687         webpage = self._webpage_read_content(
 688             full_response, url, video_id, prefix=first_bytes)
 689
 690         self.report_extraction(video_id)
 691
 692         # Is it an RSS feed?
 693         try:
 694             doc = parse_xml(webpage)
 695             if doc.tag == 'rss':
 696                 return self._extract_rss(url, video_id, doc)
 697         except compat_xml_parse_error:
 698             pass
 699
 700         # Is it a Camtasia project?
 701         camtasia_res = self._extract_camtasia(url, video_id, webpage)
 702         if camtasia_res is not None:
 703             return camtasia_res
 704
 705         # Sometimes embedded video player is hidden behind percent encoding
 706         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
 707         # Unescaping the whole page allows to handle those cases in a generic way
 708         webpage = compat_urllib_parse.unquote(webpage)
 709
 710         # it's tempting to parse this further, but you would
 711         # have to take into account all the variations like
 712         #   Video Title - Site Name
 713         #   Site Name | Video Title
 714         #   Video Title - Tagline | Site Name
 715         # and so on and so forth; it's just not practical
 716         video_title = self._html_search_regex(
 717             r'(?s)<title>(.*?)</title>', webpage, 'video title',
 718             default='video')
 719
 720         # Try to detect age limit automatically
 721         age_limit = self._rta_search(webpage)
 722         # And then there are the jokers who advertise that they use RTA,
 723         # but actually don't.
 724         AGE_LIMIT_MARKERS = [
 725             r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
 726         ]
 727         if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
 728             age_limit = 18
 729
 730         # video uploader is domain name
 731         video_uploader = self._search_regex(
 732             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
 733
 734         # Helper method
 735         def _playlist_from_matches(matches, getter=None, ie=None):
 736             urlrs = orderedSet(
 737                 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
 738                 for m in matches)
 739             return self.playlist_result(
 740                 urlrs, playlist_id=video_id, playlist_title=video_title)
 741
 742         # Look for BrightCove:
 743         bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
 744         if bc_urls:
 745             self.to_screen('Brightcove video detected.')
 746             entries = [{
 747                 '_type': 'url',
 748                 'url': smuggle_url(bc_url, {'Referer': url}),
 749                 'ie_key': 'Brightcove'
 750             } for bc_url in bc_urls]
 751
 752             return {
 753                 '_type': 'playlist',
 754                 'title': video_title,
 755                 'id': video_id,
 756                 'entries': entries,
 757             }
 758
 759         # Look for embedded (iframe) Vimeo player
 760         mobj = re.search(
 761             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
 762         if mobj:
 763             player_url = unescapeHTML(mobj.group('url'))
 764             surl = smuggle_url(player_url, {'Referer': url})
 765             return self.url_result(surl)
 766
 767         # Look for embedded (swf embed) Vimeo player
 768         mobj = re.search(
 769             r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
 770         if mobj:
 771             return self.url_result(mobj.group(1))
 772
 773         # Look for embedded YouTube player
 774         matches = re.findall(r'''(?x)
 775             (?:
 776                 <iframe[^>]+?src=|
 777                 data-video-url=|
 778                 <embed[^>]+?src=|
 779                 embedSWF\(?:\s*|
 780                 new\s+SWFObject\(
 781             )
 782             (["\'])
 783                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
 784                 (?:embed|v|p)/.+?)
 785             \1''', webpage)
 786         if matches:
 787             return _playlist_from_matches(
 788                 matches, lambda m: unescapeHTML(m[1]))
 789
 790         # Look for lazyYT YouTube embed
 791         matches = re.findall(
 792             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
 793         if matches:
 794             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
 795
 796         # Look for embedded Dailymotion player
 797         matches = re.findall(
 798             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
 799         if matches:
 800             return _playlist_from_matches(
 801                 matches, lambda m: unescapeHTML(m[1]))
 802
 803         # Look for embedded Dailymotion playlist player (#3822)
 804         m = re.search(
 805             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
 806         if m:
 807             playlists = re.findall(
 808                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
 809             if playlists:
 810                 return _playlist_from_matches(
 811                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
 812
 813         # Look for embedded Wistia player
 814         match = re.search(
 815             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
 816         if match:
 817             embed_url = self._proto_relative_url(
 818                 unescapeHTML(match.group('url')))
 819             return {
 820                 '_type': 'url_transparent',
 821                 'url': embed_url,
 822                 'ie_key': 'Wistia',
 823                 'uploader': video_uploader,
 824                 'title': video_title,
 825                 'id': video_id,
 826             }
 827
 828         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
 829         if match:
 830             return {
 831                 '_type': 'url_transparent',
 832                 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
 833                 'ie_key': 'Wistia',
 834                 'uploader': video_uploader,
 835                 'title': video_title,
 836                 'id': match.group('id')
 837             }
 838
 839         # Look for embedded blip.tv player
 840         mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
 841         if mobj:
 842             return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
 843         mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
 844         if mobj:
 845             return self.url_result(mobj.group(1), 'BlipTV')
 846
 847         # Look for embedded condenast player
 848         matches = re.findall(
 849             r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
 850             webpage)
 851         if matches:
 852             return {
 853                 '_type': 'playlist',
 854                 'entries': [{
 855                     '_type': 'url',
 856                     'ie_key': 'CondeNast',
 857                     'url': ma,
 858                 } for ma in matches],
 859                 'title': video_title,
 860                 'id': video_id,
 861             }
 862
 863         # Look for Bandcamp pages with custom domain
 864         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
 865         if mobj is not None:
 866             burl = unescapeHTML(mobj.group(1))
 867             # Don't set the extractor because it can be a track url or an album
 868             return self.url_result(burl)
 869
 870         # Look for embedded Vevo player
 871         mobj = re.search(
 872             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
 873         if mobj is not None:
 874             return self.url_result(mobj.group('url'))
 875
 876         # Look for embedded Viddler player
 877         mobj = re.search(
 878             r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
 879             webpage)
 880         if mobj is not None:
 881             return self.url_result(mobj.group('url'))
 882
 883         # Look for Ooyala videos
 884         mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
 885                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
 886         if mobj is not None:
 887             return OoyalaIE._build_url_result(mobj.group('ec'))
 888
 889         # Look for Aparat videos
 890         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
 891         if mobj is not None:
 892             return self.url_result(mobj.group(1), 'Aparat')
 893
 894         # Look for MPORA videos
 895         mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
 896         if mobj is not None:
 897             return self.url_result(mobj.group(1), 'Mpora')
 898
 899         # Look for embedded NovaMov-based player
 900         mobj = re.search(
 901             r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
 902                     (?P<url>http://(?:(?:embed|www)\.)?
 903                         (?:novamov\.com|
 904                            nowvideo\.(?:ch|sx|eu|at|ag|co)|
 905                            videoweed\.(?:es|com)|
 906                            movshare\.(?:net|sx|ag)|
 907                            divxstage\.(?:eu|net|ch|co|at|ag))
 908                         /embed\.php.+?)\1''', webpage)
 909         if mobj is not None:
 910             return self.url_result(mobj.group('url'))
 911
 912         # Look for embedded Facebook player
 913         mobj = re.search(
 914             r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
 915         if mobj is not None:
 916             return self.url_result(mobj.group('url'), 'Facebook')
 917
 918         # Look for embedded VK player
 919         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
 920         if mobj is not None:
 921             return self.url_result(mobj.group('url'), 'VK')
 922
 923         # Look for embedded ivi player
 924         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
 925         if mobj is not None:
 926             return self.url_result(mobj.group('url'), 'Ivi')
 927
 928         # Look for embedded Huffington Post player
 929         mobj = re.search(
 930             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
 931         if mobj is not None:
 932             return self.url_result(mobj.group('url'), 'HuffPost')
 933
 934         # Look for embed.ly
 935         mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
 936         if mobj is not None:
 937             return self.url_result(mobj.group('url'))
 938         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
 939         if mobj is not None:
 940             return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
 941
 942         # Look for funnyordie embed
 943         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
 944         if matches:
 945             return _playlist_from_matches(
 946                 matches, getter=unescapeHTML, ie='FunnyOrDie')
 947
 948         # Look for BBC iPlayer embed
 949         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
 950         if matches:
 951             return _playlist_from_matches(matches, ie='BBCCoUk')
 952
 953         # Look for embedded RUTV player
 954         rutv_url = RUTVIE._extract_url(webpage)
 955         if rutv_url:
 956             return self.url_result(rutv_url, 'RUTV')
 957
 958         # Look for embedded TED player
 959         mobj = re.search(
 960             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
 961         if mobj is not None:
 962             return self.url_result(mobj.group('url'), 'TED')
 963
 964         # Look for embedded Ustream videos
 965         mobj = re.search(
 966             r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
 967         if mobj is not None:
 968             return self.url_result(mobj.group('url'), 'Ustream')
 969
 970         # Look for embedded arte.tv player
 971         mobj = re.search(
 972             r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
 973             webpage)
 974         if mobj is not None:
 975             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
 976
 977         # Look for embedded smotri.com player
 978         smotri_url = SmotriIE._extract_url(webpage)
 979         if smotri_url:
 980             return self.url_result(smotri_url, 'Smotri')
 981
 982         # Look for embeded soundcloud player
 983         mobj = re.search(
 984             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
 985             webpage)
 986         if mobj is not None:
 987             url = unescapeHTML(mobj.group('url'))
 988             return self.url_result(url)
 989
 990         # Look for embedded vulture.com player
 991         mobj = re.search(
 992             r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
 993             webpage)
 994         if mobj is not None:
 995             url = unescapeHTML(mobj.group('url'))
 996             return self.url_result(url, ie='Vulture')
 997
 998         # Look for embedded mtvservices player
 999         mobj = re.search(
1000             r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1001             webpage)
1002         if mobj is not None:
1003             url = unescapeHTML(mobj.group('url'))
1004             return self.url_result(url, ie='MTVServicesEmbedded')
1005
1006         # Look for embedded yahoo player
1007         mobj = re.search(
1008             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1009             webpage)
1010         if mobj is not None:
1011             return self.url_result(mobj.group('url'), 'Yahoo')
1012
1013         # Look for embedded sbs.com.au player
1014         mobj = re.search(
1015             r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
1016             webpage)
1017         if mobj is not None:
1018             return self.url_result(mobj.group('url'), 'SBS')
1019
1020         # Look for embedded Cinchcast player
1021         mobj = re.search(
1022             r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1023             webpage)
1024         if mobj is not None:
1025             return self.url_result(mobj.group('url'), 'Cinchcast')
1026
1027         mobj = re.search(
1028             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1029             webpage)
1030         if mobj is not None:
1031             return self.url_result(mobj.group('url'), 'MLB')
1032
1033         mobj = re.search(
1034             r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1035             webpage)
1036         if mobj is not None:
1037             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1038
1039         mobj = re.search(
1040             r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1041             webpage)
1042         if mobj is not None:
1043             return self.url_result(mobj.group('url'), 'Livestream')
1044
1045         def check_video(vurl):
1046             vpath = compat_urlparse.urlparse(vurl).path
1047             vext = determine_ext(vpath)
1048             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1049
1050         def filter_video(urls):
1051             return list(filter(check_video, urls))
1052
1053         # Start with something easy: JW Player in SWFObject
1054         found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1055         if not found:
1056             # Look for gorilla-vid style embedding
1057             found = filter_video(re.findall(r'''(?sx)
1058                 (?:
1059                     jw_plugins|
1060                     JWPlayerOptions|
1061                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1062                 )
1063                 .*?file\s*:\s*["\'](.*?)["\']''', webpage))
1064         if not found:
1065             # Broaden the search a little bit
1066             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1067         if not found:
1068             # Broaden the findall a little bit: JWPlayer JS loader
1069             found = filter_video(re.findall(
1070                 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1071         if not found:
1072             # Flow player
1073             found = filter_video(re.findall(r'''(?xs)
1074                 flowplayer\("[^"]+",\s*
1075                     \{[^}]+?\}\s*,
1076                     \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1077                         ["']?url["']?\s*:\s*["']([^"']+)["']
1078             ''', webpage))
1079         if not found:
1080             # Cinerama player
1081             found = re.findall(
1082                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1083         if not found:
1084             # Try to find twitter cards info
1085             found = filter_video(re.findall(
1086                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1087         if not found:
1088             # We look for Open Graph info:
1089             # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1090             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1091             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1092             if m_video_type is not None:
1093                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1094         if not found:
1095             # HTML5 video
1096             found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1097         if not found:
1098             found = re.search(
1099                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1100                 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
1101                 webpage)
1102             if found:
1103                 new_url = found.group(1)
1104                 self.report_following_redirect(new_url)
1105                 return {
1106                     '_type': 'url',
1107                     'url': new_url,
1108                 }
1109         if not found:
1110             raise UnsupportedError(url)
1111
1112         entries = []
1113         for video_url in found:
1114             video_url = compat_urlparse.urljoin(url, video_url)
1115             video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1116
1117             # Sometimes, jwplayer extraction will result in a YouTube URL
1118             if YoutubeIE.suitable(video_url):
1119                 entries.append(self.url_result(video_url, 'Youtube'))
1120                 continue
1121
1122             # here's a fun little line of code for you:
1123             video_id = os.path.splitext(video_id)[0]
1124
1125             entries.append({
1126                 'id': video_id,
1127                 'url': video_url,
1128                 'uploader': video_uploader,
1129                 'title': video_title,
1130                 'age_limit': age_limit,
1131             })
1132
1133         if len(entries) == 1:
1134             return entries[0]
1135         else:
1136             for num, e in enumerate(entries, start=1):
1137                 e['title'] = '%s (%d)' % (e['title'], num)
1138             return {
1139                 '_type': 'playlist',
1140                 'entries': entries,
1141             }