youtube_dl/extractor/bbc.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import xml.etree.ElementTree
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     ExtractorError,
  10     float_or_none,
  11     int_or_none,
  12     parse_duration,
  13     parse_iso8601,
  14     unescapeHTML,
  15 )
  16 from ..compat import compat_HTTPError
  17
  18
  19 class BBCCoUkIE(InfoExtractor):
  20     IE_NAME = 'bbc.co.uk'
  21     IE_DESC = 'BBC iPlayer'
  22     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
  23
  24     _MEDIASELECTOR_URLS = [
  25         # Provides HQ HLS streams with even better quality that pc mediaset but fails
  26         # with geolocation in some cases when it's even not geo restricted at all (e.g.
  27         # http://www.bbc.co.uk/programmes/b06bp7lf)
  28         'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
  29         'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
  30     ]
  31
  32     _TESTS = [
  33         {
  34             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
  35             'info_dict': {
  36                 'id': 'b039d07m',
  37                 'ext': 'flv',
  38                 'title': 'Kaleidoscope, Leonard Cohen',
  39                 'description': 'The Canadian poet and songwriter reflects on his musical career.',
  40                 'duration': 1740,
  41             },
  42             'params': {
  43                 # rtmp download
  44                 'skip_download': True,
  45             }
  46         },
  47         {
  48             'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
  49             'info_dict': {
  50                 'id': 'b00yng1d',
  51                 'ext': 'flv',
  52                 'title': 'The Man in Black: Series 3: The Printed Name',
  53                 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
  54                 'duration': 1800,
  55             },
  56             'params': {
  57                 # rtmp download
  58                 'skip_download': True,
  59             },
  60             'skip': 'Episode is no longer available on BBC iPlayer Radio',
  61         },
  62         {
  63             'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
  64             'info_dict': {
  65                 'id': 'b00yng1d',
  66                 'ext': 'flv',
  67                 'title': 'The Voice UK: Series 3: Blind Auditions 5',
  68                 'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
  69                 'duration': 5100,
  70             },
  71             'params': {
  72                 # rtmp download
  73                 'skip_download': True,
  74             },
  75             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  76         },
  77         {
  78             'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
  79             'info_dict': {
  80                 'id': 'b03k3pb7',
  81                 'ext': 'flv',
  82                 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
  83                 'description': '2. Invasion',
  84                 'duration': 3600,
  85             },
  86             'params': {
  87                 # rtmp download
  88                 'skip_download': True,
  89             },
  90             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  91         }, {
  92             'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
  93             'info_dict': {
  94                 'id': 'b04v209v',
  95                 'ext': 'flv',
  96                 'title': 'Pete Tong, The Essential New Tune Special',
  97                 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
  98                 'duration': 10800,
  99             },
 100             'params': {
 101                 # rtmp download
 102                 'skip_download': True,
 103             }
 104         }, {
 105             'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
 106             'note': 'Audio',
 107             'info_dict': {
 108                 'id': 'p02frcch',
 109                 'ext': 'flv',
 110                 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
 111                 'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
 112                 'duration': 3507,
 113             },
 114             'params': {
 115                 # rtmp download
 116                 'skip_download': True,
 117             }
 118         }, {
 119             'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
 120             'note': 'Video',
 121             'info_dict': {
 122                 'id': 'p025c103',
 123                 'ext': 'flv',
 124                 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
 125                 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
 126                 'duration': 226,
 127             },
 128             'params': {
 129                 # rtmp download
 130                 'skip_download': True,
 131             }
 132         }, {
 133             'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
 134             'info_dict': {
 135                 'id': 'p02n76xf',
 136                 'ext': 'flv',
 137                 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
 138                 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
 139                 'duration': 3540,
 140             },
 141             'params': {
 142                 # rtmp download
 143                 'skip_download': True,
 144             },
 145             'skip': 'geolocation',
 146         }, {
 147             'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
 148             'info_dict': {
 149                 'id': 'b05zmgw1',
 150                 'ext': 'flv',
 151                 'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.',
 152                 'title': 'Royal Academy Summer Exhibition',
 153                 'duration': 3540,
 154             },
 155             'params': {
 156                 # rtmp download
 157                 'skip_download': True,
 158             },
 159             'skip': 'geolocation',
 160         }, {
 161             # iptv-all mediaset fails with geolocation however there is no geo restriction
 162             # for this programme at all
 163             'url': 'http://www.bbc.co.uk/programmes/b06bp7lf',
 164             'info_dict': {
 165                 'id': 'b06bp7kf',
 166                 'ext': 'flv',
 167                 'title': "Annie Mac's Friday Night, B.Traits sits in for Annie",
 168                 'description': 'B.Traits sits in for Annie Mac with a Mini-Mix from Disclosure.',
 169                 'duration': 10800,
 170             },
 171             'params': {
 172                 # rtmp download
 173                 'skip_download': True,
 174             },
 175         }, {
 176             'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
 177             'only_matching': True,
 178         }, {
 179             'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
 180             'only_matching': True,
 181         }, {
 182             'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
 183             'only_matching': True,
 184         }
 185     ]
 186
 187     class MediaSelectionError(Exception):
 188         def __init__(self, id):
 189             self.id = id
 190
 191     def _extract_asx_playlist(self, connection, programme_id):
 192         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
 193         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
 194
 195     def _extract_connection(self, connection, programme_id):
 196         formats = []
 197         protocol = connection.get('protocol')
 198         supplier = connection.get('supplier')
 199         if protocol == 'http':
 200             href = connection.get('href')
 201             transfer_format = connection.get('transferFormat')
 202             # ASX playlist
 203             if supplier == 'asx':
 204                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
 205                     formats.append({
 206                         'url': ref,
 207                         'format_id': 'ref%s_%s' % (i, supplier),
 208                     })
 209             # Skip DASH until supported
 210             elif transfer_format == 'dash':
 211                 pass
 212             elif transfer_format == 'hls':
 213                 m3u8_formats = self._extract_m3u8_formats(
 214                     href, programme_id, ext='mp4', entry_protocol='m3u8_native',
 215                     m3u8_id=supplier, fatal=False)
 216                 if m3u8_formats:
 217                     formats.extend(m3u8_formats)
 218             # Direct link
 219             else:
 220                 formats.append({
 221                     'url': href,
 222                     'format_id': supplier,
 223                 })
 224         elif protocol == 'rtmp':
 225             application = connection.get('application', 'ondemand')
 226             auth_string = connection.get('authString')
 227             identifier = connection.get('identifier')
 228             server = connection.get('server')
 229             formats.append({
 230                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 231                 'play_path': identifier,
 232                 'app': '%s?%s' % (application, auth_string),
 233                 'page_url': 'http://www.bbc.co.uk',
 234                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 235                 'rtmp_live': False,
 236                 'ext': 'flv',
 237                 'format_id': supplier,
 238             })
 239         return formats
 240
 241     def _extract_items(self, playlist):
 242         return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
 243
 244     def _extract_medias(self, media_selection):
 245         error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
 246         if error is not None:
 247             raise BBCCoUkIE.MediaSelectionError(error.get('id'))
 248         return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
 249
 250     def _extract_connections(self, media):
 251         return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
 252
 253     def _extract_video(self, media, programme_id):
 254         formats = []
 255         vbr = int_or_none(media.get('bitrate'))
 256         vcodec = media.get('encoding')
 257         service = media.get('service')
 258         width = int_or_none(media.get('width'))
 259         height = int_or_none(media.get('height'))
 260         file_size = int_or_none(media.get('media_file_size'))
 261         for connection in self._extract_connections(media):
 262             conn_formats = self._extract_connection(connection, programme_id)
 263             for format in conn_formats:
 264                 format.update({
 265                     'format_id': '%s_%s' % (service, format['format_id']),
 266                     'width': width,
 267                     'height': height,
 268                     'vbr': vbr,
 269                     'vcodec': vcodec,
 270                     'filesize': file_size,
 271                 })
 272             formats.extend(conn_formats)
 273         return formats
 274
 275     def _extract_audio(self, media, programme_id):
 276         formats = []
 277         abr = int_or_none(media.get('bitrate'))
 278         acodec = media.get('encoding')
 279         service = media.get('service')
 280         for connection in self._extract_connections(media):
 281             conn_formats = self._extract_connection(connection, programme_id)
 282             for format in conn_formats:
 283                 format.update({
 284                     'format_id': '%s_%s' % (service, format['format_id']),
 285                     'abr': abr,
 286                     'acodec': acodec,
 287                 })
 288             formats.extend(conn_formats)
 289         return formats
 290
 291     def _get_subtitles(self, media, programme_id):
 292         subtitles = {}
 293         for connection in self._extract_connections(media):
 294             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 295             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 296             subtitles[lang] = [
 297                 {
 298                     'url': connection.get('href'),
 299                     'ext': 'ttml',
 300                 },
 301             ]
 302         return subtitles
 303
 304     def _raise_extractor_error(self, media_selection_error):
 305         raise ExtractorError(
 306             '%s returned error: %s' % (self.IE_NAME, media_selection_error.id),
 307             expected=True)
 308
 309     def _download_media_selector(self, programme_id):
 310         last_exception = None
 311         for mediaselector_url in self._MEDIASELECTOR_URLS:
 312             try:
 313                 return self._download_media_selector_url(
 314                     mediaselector_url % programme_id, programme_id)
 315             except BBCCoUkIE.MediaSelectionError as e:
 316                 if e.id in ('notukerror', 'geolocation'):
 317                     last_exception = e
 318                     continue
 319                 self._raise_extractor_error(e)
 320         self._raise_extractor_error(last_exception)
 321
 322     def _download_media_selector_url(self, url, programme_id=None):
 323         try:
 324             media_selection = self._download_xml(
 325                 url, programme_id, 'Downloading media selection XML')
 326         except ExtractorError as ee:
 327             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 328                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8'))
 329             else:
 330                 raise
 331         return self._process_media_selector(media_selection, programme_id)
 332
 333     def _process_media_selector(self, media_selection, programme_id):
 334         formats = []
 335         subtitles = None
 336
 337         for media in self._extract_medias(media_selection):
 338             kind = media.get('kind')
 339             if kind == 'audio':
 340                 formats.extend(self._extract_audio(media, programme_id))
 341             elif kind == 'video':
 342                 formats.extend(self._extract_video(media, programme_id))
 343             elif kind == 'captions':
 344                 subtitles = self.extract_subtitles(media, programme_id)
 345         return formats, subtitles
 346
 347     def _download_playlist(self, playlist_id):
 348         try:
 349             playlist = self._download_json(
 350                 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
 351                 playlist_id, 'Downloading playlist JSON')
 352
 353             version = playlist.get('defaultAvailableVersion')
 354             if version:
 355                 smp_config = version['smpConfig']
 356                 title = smp_config['title']
 357                 description = smp_config['summary']
 358                 for item in smp_config['items']:
 359                     kind = item['kind']
 360                     if kind != 'programme' and kind != 'radioProgramme':
 361                         continue
 362                     programme_id = item.get('vpid')
 363                     duration = int_or_none(item.get('duration'))
 364                     formats, subtitles = self._download_media_selector(programme_id)
 365                 return programme_id, title, description, duration, formats, subtitles
 366         except ExtractorError as ee:
 367             if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
 368                 raise
 369
 370         # fallback to legacy playlist
 371         return self._process_legacy_playlist(playlist_id)
 372
 373     def _process_legacy_playlist_url(self, url, display_id):
 374         playlist = self._download_legacy_playlist_url(url, display_id)
 375         return self._extract_from_legacy_playlist(playlist, display_id)
 376
 377     def _process_legacy_playlist(self, playlist_id):
 378         return self._process_legacy_playlist_url(
 379             'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
 380
 381     def _download_legacy_playlist_url(self, url, playlist_id=None):
 382         return self._download_xml(
 383             url, playlist_id, 'Downloading legacy playlist XML')
 384
 385     def _extract_from_legacy_playlist(self, playlist, playlist_id):
 386         no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
 387         if no_items is not None:
 388             reason = no_items.get('reason')
 389             if reason == 'preAvailability':
 390                 msg = 'Episode %s is not yet available' % playlist_id
 391             elif reason == 'postAvailability':
 392                 msg = 'Episode %s is no longer available' % playlist_id
 393             elif reason == 'noMedia':
 394                 msg = 'Episode %s is not currently available' % playlist_id
 395             else:
 396                 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
 397             raise ExtractorError(msg, expected=True)
 398
 399         for item in self._extract_items(playlist):
 400             kind = item.get('kind')
 401             if kind != 'programme' and kind != 'radioProgramme':
 402                 continue
 403             title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
 404             description_el = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary')
 405             description = description_el.text if description_el else None
 406
 407             def get_programme_id(item):
 408                 def get_from_attributes(item):
 409                     for p in('identifier', 'group'):
 410                         value = item.get(p)
 411                         if value and re.match(r'^[pb][\da-z]{7}$', value):
 412                             return value
 413                 get_from_attributes(item)
 414                 mediator = item.find('./{http://bbc.co.uk/2008/emp/playlist}mediator')
 415                 if mediator is not None:
 416                     return get_from_attributes(mediator)
 417
 418             programme_id = get_programme_id(item)
 419             duration = int_or_none(item.get('duration'))
 420             # TODO: programme_id can be None and media items can be incorporated right inside
 421             # playlist's item (e.g. http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
 422             # as f4m and m3u8
 423             formats, subtitles = self._download_media_selector(programme_id)
 424
 425         return programme_id, title, description, duration, formats, subtitles
 426
 427     def _real_extract(self, url):
 428         group_id = self._match_id(url)
 429
 430         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 431
 432         programme_id = None
 433
 434         tviplayer = self._search_regex(
 435             r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
 436             webpage, 'player', default=None)
 437
 438         if tviplayer:
 439             player = self._parse_json(tviplayer, group_id).get('player', {})
 440             duration = int_or_none(player.get('duration'))
 441             programme_id = player.get('vpid')
 442
 443         if not programme_id:
 444             programme_id = self._search_regex(
 445                 r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
 446
 447         if programme_id:
 448             formats, subtitles = self._download_media_selector(programme_id)
 449             title = self._og_search_title(webpage)
 450             description = self._search_regex(
 451                 r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
 452                 webpage, 'description', fatal=False)
 453         else:
 454             programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
 455
 456         self._sort_formats(formats)
 457
 458         return {
 459             'id': programme_id,
 460             'title': title,
 461             'description': description,
 462             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 463             'duration': duration,
 464             'formats': formats,
 465             'subtitles': subtitles,
 466         }
 467
 468
 469 class BBCIE(BBCCoUkIE):
 470     IE_NAME = 'bbc'
 471     IE_DESC = 'BBC'
 472     _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
 473
 474     _MEDIASELECTOR_URLS = [
 475         # Provides more formats, namely direct mp4 links, but fails on some videos with
 476         # notukerror for non UK (?) users (e.g.
 477         # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
 478         'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
 479         # Provides fewer formats, but works everywhere for everybody (hopefully)
 480         'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
 481     ]
 482
 483     _TESTS = [{
 484         # article with multiple videos embedded with data-media-meta containing
 485         # playlist.sxml, externalId and no direct video links
 486         'url': 'http://www.bbc.com/news/world-europe-32668511',
 487         'info_dict': {
 488             'id': 'world-europe-32668511',
 489             'title': 'Russia stages massive WW2 parade despite Western boycott',
 490             'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
 491         },
 492         'playlist_count': 2,
 493     }, {
 494         # article with multiple videos embedded with data-media-meta (more videos)
 495         'url': 'http://www.bbc.com/news/business-28299555',
 496         'info_dict': {
 497             'id': 'business-28299555',
 498             'title': 'Farnborough Airshow: Video highlights',
 499             'description': 'BBC reports and video highlights at the Farnborough Airshow.',
 500         },
 501         'playlist_count': 9,
 502         'skip': 'Save time',
 503     }, {
 504         # article with multiple videos embedded with `new SMP()`
 505         'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
 506         'info_dict': {
 507             'id': '3662a707-0af9-3149-963f-47bea720b460',
 508             'title': 'BBC Blogs - Adam Curtis - BUGGER',
 509         },
 510         'playlist_count': 18,
 511     }, {
 512         # single video embedded with mediaAssetPage.init()
 513         'url': 'http://www.bbc.com/news/world-europe-32041533',
 514         'info_dict': {
 515             'id': 'p02mprgb',
 516             'ext': 'mp4',
 517             'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
 518             'duration': 47,
 519             'timestamp': 1427219242,
 520             'upload_date': '20150324',
 521         },
 522         'params': {
 523             # rtmp download
 524             'skip_download': True,
 525         }
 526     }, {
 527         # article with single video embedded with data-media-meta containing
 528         # direct video links (for now these are extracted) and playlist.xml (with
 529         # media items as f4m and m3u8 - currently unsupported)
 530         'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
 531         'info_dict': {
 532             'id': '150615_telabyad_kentin_cogu',
 533             'ext': 'mp4',
 534             'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
 535             'duration': 47,
 536             'timestamp': 1434397334,
 537             'upload_date': '20150615',
 538         },
 539         'params': {
 540             'skip_download': True,
 541         }
 542     }, {
 543         # single video embedded with mediaAssetPage.init() (regional section)
 544         'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
 545         'info_dict': {
 546             'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
 547             'ext': 'mp4',
 548             'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
 549             'duration': 87,
 550             'timestamp': 1434713142,
 551             'upload_date': '20150619',
 552         },
 553         'params': {
 554             'skip_download': True,
 555         }
 556     }, {
 557         # single video from video playlist embedded with vxp-playlist-data JSON
 558         'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
 559         'info_dict': {
 560             'id': 'p02w6qjc',
 561             'ext': 'mp4',
 562             'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
 563             'duration': 56,
 564         },
 565         'params': {
 566             'skip_download': True,
 567         }
 568     }, {
 569         # single video story with digitalData
 570         'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
 571         'info_dict': {
 572             'id': 'p02q6gc4',
 573             'ext': 'flv',
 574             'title': 'Sri Lanka’s spicy secret',
 575             'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
 576             'timestamp': 1437674293,
 577             'upload_date': '20150723',
 578         },
 579         'params': {
 580             # rtmp download
 581             'skip_download': True,
 582         }
 583     }, {
 584         # single video story without digitalData
 585         'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
 586         'info_dict': {
 587             'id': 'p018zqqg',
 588             'ext': 'mp4',
 589             'title': 'Hyundai Santa Fe Sport: Rock star',
 590             'description': 'md5:b042a26142c4154a6e472933cf20793d',
 591             'timestamp': 1368473503,
 592             'upload_date': '20130513',
 593         },
 594         'params': {
 595             # rtmp download
 596             'skip_download': True,
 597         }
 598     }, {
 599         # single video with playlist.sxml URL
 600         'url': 'http://www.bbc.com/sport/0/football/33653409',
 601         'info_dict': {
 602             'id': 'p02xycnp',
 603             'ext': 'mp4',
 604             'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
 605             'description': 'md5:398fca0e2e701c609d726e034fa1fc89',
 606             'duration': 140,
 607         },
 608         'params': {
 609             # rtmp download
 610             'skip_download': True,
 611         }
 612     }, {
 613         # article with multiple videos embedded with playlist.sxml
 614         'url': 'http://www.bbc.com/sport/0/football/34475836',
 615         'info_dict': {
 616             'id': '34475836',
 617             'title': 'What Liverpool can expect from Klopp',
 618         },
 619         'playlist_count': 3,
 620     }, {
 621         # single video with playlist URL from weather section
 622         'url': 'http://www.bbc.com/weather/features/33601775',
 623         'only_matching': True,
 624     }, {
 625         # custom redirection to www.bbc.com
 626         'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
 627         'only_matching': True,
 628     }]
 629
 630     @classmethod
 631     def suitable(cls, url):
 632         return False if BBCCoUkIE.suitable(url) else super(BBCIE, cls).suitable(url)
 633
 634     def _extract_from_media_meta(self, media_meta, video_id):
 635         # Direct links to media in media metadata (e.g.
 636         # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
 637         # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
 638         source_files = media_meta.get('sourceFiles')
 639         if source_files:
 640             return [{
 641                 'url': f['url'],
 642                 'format_id': format_id,
 643                 'ext': f.get('encoding'),
 644                 'tbr': float_or_none(f.get('bitrate'), 1000),
 645                 'filesize': int_or_none(f.get('filesize')),
 646             } for format_id, f in source_files.items() if f.get('url')], []
 647
 648         programme_id = media_meta.get('externalId')
 649         if programme_id:
 650             return self._download_media_selector(programme_id)
 651
 652         # Process playlist.sxml as legacy playlist
 653         href = media_meta.get('href')
 654         if href:
 655             playlist = self._download_legacy_playlist_url(href)
 656             _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
 657             return formats, subtitles
 658
 659         return [], []
 660
 661     def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
 662         programme_id, title, description, duration, formats, subtitles = \
 663             self._process_legacy_playlist_url(url, playlist_id)
 664         self._sort_formats(formats)
 665         return {
 666             'id': programme_id,
 667             'title': title,
 668             'description': description,
 669             'duration': duration,
 670             'timestamp': timestamp,
 671             'formats': formats,
 672             'subtitles': subtitles,
 673         }
 674
 675     def _real_extract(self, url):
 676         playlist_id = self._match_id(url)
 677
 678         webpage = self._download_webpage(url, playlist_id)
 679
 680         timestamp = parse_iso8601(self._search_regex(
 681             [r'"datePublished":\s*"([^"]+)',
 682              r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
 683              r'itemprop="datePublished"[^>]+datetime="([^"]+)"'],
 684             webpage, 'date', default=None))
 685
 686         # article with multiple videos embedded with playlist.sxml (e.g.
 687         # http://www.bbc.com/sport/0/football/34475836)
 688         playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
 689         if playlists:
 690             entries = [
 691                 self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
 692                 for playlist_url in playlists]
 693             playlist_title = self._og_search_title(webpage)
 694             playlist_description = self._og_search_description(webpage, default=None)
 695             return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 696
 697         # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
 698         programme_id = self._search_regex(
 699             [r'data-video-player-vpid="([\da-z]{8})"',
 700              r'<param[^>]+name="externalIdentifier"[^>]+value="([\da-z]{8})"'],
 701             webpage, 'vpid', default=None)
 702
 703         duration = None
 704         if not programme_id:
 705             # single video in news article embedded with data-playable (e.g.
 706             # http://www.bbc.com/news/world-us-canada-34473351)
 707             data_playable = self._parse_json(
 708                 unescapeHTML(self._search_regex(
 709                     r'data-playable="({.+?})"', webpage, 'data playable', default='{}')),
 710                 programme_id, fatal=False)
 711             if data_playable:
 712                 # data-playable has video vpid in settings.playlistObject.items (e.g.
 713                 # http://www.bbc.com/news/world-us-canada-34473351)
 714                 settings = data_playable.get('settings', {})
 715                 if settings:
 716                     playlist_object = settings.get('playlistObject', {})
 717                     if playlist_object:
 718                         items = playlist_object.get('items')
 719                         if items and isinstance(items, list):
 720                             duration = int_or_none(items[0].get('duration'))
 721                             programme_id = items[0].get('vpid')
 722                 if not programme_id:
 723                     # data-playable has no vpid but has a playlist.sxml URLs
 724                     # in otherSettings.playlist (e.g.
 725                     # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
 726                     playlist = data_playable.get('otherSettings', {}).get('playlist', {})
 727                     if playlist:
 728                         return self._extract_from_playlist_sxml(
 729                             playlist.get('progressiveDownloadUrl'), playlist_id, timestamp)
 730
 731         if programme_id:
 732             formats, subtitles = self._download_media_selector(programme_id)
 733             self._sort_formats(formats)
 734             # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
 735             digital_data = self._parse_json(
 736                 self._search_regex(
 737                     r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
 738                 programme_id, fatal=False)
 739             page_info = digital_data.get('page', {}).get('pageInfo', {})
 740             title = page_info.get('pageName') or self._og_search_title(webpage)
 741             description = page_info.get('description') or self._og_search_description(webpage)
 742             timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
 743             return {
 744                 'id': programme_id,
 745                 'title': title,
 746                 'description': description,
 747                 'timestamp': timestamp,
 748                 'duration': duration,
 749                 'formats': formats,
 750                 'subtitles': subtitles,
 751             }
 752
 753         playlist_title = self._html_search_regex(
 754             r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'playlist title')
 755         playlist_description = self._og_search_description(webpage, default=None)
 756
 757         def extract_all(pattern):
 758             return list(filter(None, map(
 759                 lambda s: self._parse_json(s, playlist_id, fatal=False),
 760                 re.findall(pattern, webpage))))
 761
 762         # Multiple video article (e.g.
 763         # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
 764         EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+[\da-z]{8}(?:\b[^"]+)?'
 765         entries = []
 766         for match in extract_all(r'new\s+SMP\(({.+?})\)'):
 767             embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
 768             if embed_url and re.match(EMBED_URL, embed_url):
 769                 entries.append(embed_url)
 770         entries.extend(re.findall(
 771             r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
 772         if entries:
 773             return self.playlist_result(
 774                 [self.url_result(entry, 'BBCCoUk') for entry in entries],
 775                 playlist_id, playlist_title, playlist_description)
 776
 777         # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
 778         medias = extract_all(r"data-media-meta='({[^']+})'")
 779
 780         if not medias:
 781             # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
 782             media_asset = self._search_regex(
 783                 r'mediaAssetPage\.init\(\s*({.+?}), "/',
 784                 webpage, 'media asset', default=None)
 785             if media_asset:
 786                 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
 787                 medias = []
 788                 for video in media_asset_page.get('videos', {}).values():
 789                     medias.extend(video.values())
 790
 791         if not medias:
 792             # Multiple video playlist with single `now playing` entry (e.g.
 793             # http://www.bbc.com/news/video_and_audio/must_see/33767813)
 794             vxp_playlist = self._parse_json(
 795                 self._search_regex(
 796                     r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>',
 797                     webpage, 'playlist data'),
 798                 playlist_id)
 799             playlist_medias = []
 800             for item in vxp_playlist:
 801                 media = item.get('media')
 802                 if not media:
 803                     continue
 804                 playlist_medias.append(media)
 805                 # Download single video if found media with asset id matching the video id from URL
 806                 if item.get('advert', {}).get('assetId') == playlist_id:
 807                     medias = [media]
 808                     break
 809             # Fallback to the whole playlist
 810             if not medias:
 811                 medias = playlist_medias
 812
 813         entries = []
 814         for num, media_meta in enumerate(medias, start=1):
 815             formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
 816             if not formats:
 817                 continue
 818             self._sort_formats(formats)
 819
 820             video_id = media_meta.get('externalId')
 821             if not video_id:
 822                 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
 823
 824             title = media_meta.get('caption')
 825             if not title:
 826                 title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
 827
 828             duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
 829
 830             images = []
 831             for image in media_meta.get('images', {}).values():
 832                 images.extend(image.values())
 833             if 'image' in media_meta:
 834                 images.append(media_meta['image'])
 835
 836             thumbnails = [{
 837                 'url': image.get('href'),
 838                 'width': int_or_none(image.get('width')),
 839                 'height': int_or_none(image.get('height')),
 840             } for image in images]
 841
 842             entries.append({
 843                 'id': video_id,
 844                 'title': title,
 845                 'thumbnails': thumbnails,
 846                 'duration': duration,
 847                 'timestamp': timestamp,
 848                 'formats': formats,
 849                 'subtitles': subtitles,
 850             })
 851
 852         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)