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     _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
  33     _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
  34
  35     _NAMESPACES = (
  36         _MEDIASELECTION_NS,
  37         _EMP_PLAYLIST_NS,
  38     )
  39
  40     _TESTS = [
  41         {
  42             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
  43             'info_dict': {
  44                 'id': 'b039d07m',
  45                 'ext': 'flv',
  46                 'title': 'Kaleidoscope, Leonard Cohen',
  47                 'description': 'The Canadian poet and songwriter reflects on his musical career.',
  48                 'duration': 1740,
  49             },
  50             'params': {
  51                 # rtmp download
  52                 'skip_download': True,
  53             }
  54         },
  55         {
  56             'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
  57             'info_dict': {
  58                 'id': 'b00yng1d',
  59                 'ext': 'flv',
  60                 'title': 'The Man in Black: Series 3: The Printed Name',
  61                 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
  62                 'duration': 1800,
  63             },
  64             'params': {
  65                 # rtmp download
  66                 'skip_download': True,
  67             },
  68             'skip': 'Episode is no longer available on BBC iPlayer Radio',
  69         },
  70         {
  71             'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
  72             'info_dict': {
  73                 'id': 'b00yng1d',
  74                 'ext': 'flv',
  75                 'title': 'The Voice UK: Series 3: Blind Auditions 5',
  76                 '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.",
  77                 'duration': 5100,
  78             },
  79             'params': {
  80                 # rtmp download
  81                 'skip_download': True,
  82             },
  83             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  84         },
  85         {
  86             'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
  87             'info_dict': {
  88                 'id': 'b03k3pb7',
  89                 'ext': 'flv',
  90                 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
  91                 'description': '2. Invasion',
  92                 'duration': 3600,
  93             },
  94             'params': {
  95                 # rtmp download
  96                 'skip_download': True,
  97             },
  98             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  99         }, {
 100             'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
 101             'info_dict': {
 102                 'id': 'b04v209v',
 103                 'ext': 'flv',
 104                 'title': 'Pete Tong, The Essential New Tune Special',
 105                 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
 106                 'duration': 10800,
 107             },
 108             'params': {
 109                 # rtmp download
 110                 'skip_download': True,
 111             }
 112         }, {
 113             'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
 114             'note': 'Audio',
 115             'info_dict': {
 116                 'id': 'p02frcch',
 117                 'ext': 'flv',
 118                 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
 119                 'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
 120                 'duration': 3507,
 121             },
 122             'params': {
 123                 # rtmp download
 124                 'skip_download': True,
 125             }
 126         }, {
 127             'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
 128             'note': 'Video',
 129             'info_dict': {
 130                 'id': 'p025c103',
 131                 'ext': 'flv',
 132                 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
 133                 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
 134                 'duration': 226,
 135             },
 136             'params': {
 137                 # rtmp download
 138                 'skip_download': True,
 139             }
 140         }, {
 141             'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
 142             'info_dict': {
 143                 'id': 'p02n76xf',
 144                 'ext': 'flv',
 145                 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
 146                 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
 147                 'duration': 3540,
 148             },
 149             'params': {
 150                 # rtmp download
 151                 'skip_download': True,
 152             },
 153             'skip': 'geolocation',
 154         }, {
 155             'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
 156             'info_dict': {
 157                 'id': 'b05zmgw1',
 158                 'ext': 'flv',
 159                 '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.',
 160                 'title': 'Royal Academy Summer Exhibition',
 161                 'duration': 3540,
 162             },
 163             'params': {
 164                 # rtmp download
 165                 'skip_download': True,
 166             },
 167             'skip': 'geolocation',
 168         }, {
 169             # iptv-all mediaset fails with geolocation however there is no geo restriction
 170             # for this programme at all
 171             'url': 'http://www.bbc.co.uk/programmes/b06bp7lf',
 172             'info_dict': {
 173                 'id': 'b06bp7kf',
 174                 'ext': 'flv',
 175                 'title': "Annie Mac's Friday Night, B.Traits sits in for Annie",
 176                 'description': 'B.Traits sits in for Annie Mac with a Mini-Mix from Disclosure.',
 177                 'duration': 10800,
 178             },
 179             'params': {
 180                 # rtmp download
 181                 'skip_download': True,
 182             },
 183         }, {
 184             'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
 185             'only_matching': True,
 186         }, {
 187             'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
 188             'only_matching': True,
 189         }, {
 190             'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
 191             'only_matching': True,
 192         }
 193     ]
 194
 195     class MediaSelectionError(Exception):
 196         def __init__(self, id):
 197             self.id = id
 198
 199     def _extract_asx_playlist(self, connection, programme_id):
 200         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
 201         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
 202
 203     def _extract_connection(self, connection, programme_id):
 204         formats = []
 205         kind = connection.get('kind')
 206         protocol = connection.get('protocol')
 207         supplier = connection.get('supplier')
 208         if protocol == 'http':
 209             href = connection.get('href')
 210             transfer_format = connection.get('transferFormat')
 211             # ASX playlist
 212             if supplier == 'asx':
 213                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
 214                     formats.append({
 215                         'url': ref,
 216                         'format_id': 'ref%s_%s' % (i, supplier),
 217                     })
 218             # Skip DASH until supported
 219             elif transfer_format == 'dash':
 220                 pass
 221             elif transfer_format == 'hls':
 222                 m3u8_formats = self._extract_m3u8_formats(
 223                     href, programme_id, ext='mp4', entry_protocol='m3u8_native',
 224                     m3u8_id=supplier, fatal=False)
 225                 if m3u8_formats:
 226                     formats.extend(m3u8_formats)
 227             # Direct link
 228             else:
 229                 formats.append({
 230                     'url': href,
 231                     'format_id': supplier or kind or protocol,
 232                 })
 233         elif protocol == 'rtmp':
 234             application = connection.get('application', 'ondemand')
 235             auth_string = connection.get('authString')
 236             identifier = connection.get('identifier')
 237             server = connection.get('server')
 238             formats.append({
 239                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 240                 'play_path': identifier,
 241                 'app': '%s?%s' % (application, auth_string),
 242                 'page_url': 'http://www.bbc.co.uk',
 243                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 244                 'rtmp_live': False,
 245                 'ext': 'flv',
 246                 'format_id': supplier,
 247             })
 248         return formats
 249
 250     def _extract_items(self, playlist):
 251         return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
 252
 253     def _findall_ns(self, element, xpath):
 254         elements = []
 255         for ns in self._NAMESPACES:
 256             elements.extend(element.findall(xpath % ns))
 257         return elements
 258
 259     def _extract_medias(self, media_selection):
 260         error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS)
 261         if error is None:
 262             media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS)
 263         if error is not None:
 264             raise BBCCoUkIE.MediaSelectionError(error.get('id'))
 265         return self._findall_ns(media_selection, './{%s}media')
 266
 267     def _extract_connections(self, media):
 268         return self._findall_ns(media, './{%s}connection')
 269
 270     def _extract_video(self, media, programme_id):
 271         formats = []
 272         vbr = int_or_none(media.get('bitrate'))
 273         vcodec = media.get('encoding')
 274         service = media.get('service')
 275         width = int_or_none(media.get('width'))
 276         height = int_or_none(media.get('height'))
 277         file_size = int_or_none(media.get('media_file_size'))
 278         for connection in self._extract_connections(media):
 279             conn_formats = self._extract_connection(connection, programme_id)
 280             for format in conn_formats:
 281                 format.update({
 282                     'width': width,
 283                     'height': height,
 284                     'vbr': vbr,
 285                     'vcodec': vcodec,
 286                     'filesize': file_size,
 287                 })
 288                 if service:
 289                     format['format_id'] = '%s_%s' % (service, format['format_id'])
 290             formats.extend(conn_formats)
 291         return formats
 292
 293     def _extract_audio(self, media, programme_id):
 294         formats = []
 295         abr = int_or_none(media.get('bitrate'))
 296         acodec = media.get('encoding')
 297         service = media.get('service')
 298         for connection in self._extract_connections(media):
 299             conn_formats = self._extract_connection(connection, programme_id)
 300             for format in conn_formats:
 301                 format.update({
 302                     'format_id': '%s_%s' % (service, format['format_id']),
 303                     'abr': abr,
 304                     'acodec': acodec,
 305                 })
 306             formats.extend(conn_formats)
 307         return formats
 308
 309     def _get_subtitles(self, media, programme_id):
 310         subtitles = {}
 311         for connection in self._extract_connections(media):
 312             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 313             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 314             subtitles[lang] = [
 315                 {
 316                     'url': connection.get('href'),
 317                     'ext': 'ttml',
 318                 },
 319             ]
 320         return subtitles
 321
 322     def _raise_extractor_error(self, media_selection_error):
 323         raise ExtractorError(
 324             '%s returned error: %s' % (self.IE_NAME, media_selection_error.id),
 325             expected=True)
 326
 327     def _download_media_selector(self, programme_id):
 328         last_exception = None
 329         for mediaselector_url in self._MEDIASELECTOR_URLS:
 330             try:
 331                 return self._download_media_selector_url(
 332                     mediaselector_url % programme_id, programme_id)
 333             except BBCCoUkIE.MediaSelectionError as e:
 334                 if e.id in ('notukerror', 'geolocation'):
 335                     last_exception = e
 336                     continue
 337                 self._raise_extractor_error(e)
 338         self._raise_extractor_error(last_exception)
 339
 340     def _download_media_selector_url(self, url, programme_id=None):
 341         try:
 342             media_selection = self._download_xml(
 343                 url, programme_id, 'Downloading media selection XML')
 344         except ExtractorError as ee:
 345             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 346                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8'))
 347             else:
 348                 raise
 349         return self._process_media_selector(media_selection, programme_id)
 350
 351     def _process_media_selector(self, media_selection, programme_id):
 352         formats = []
 353         subtitles = None
 354
 355         for media in self._extract_medias(media_selection):
 356             kind = media.get('kind')
 357             if kind == 'audio':
 358                 formats.extend(self._extract_audio(media, programme_id))
 359             elif kind == 'video':
 360                 formats.extend(self._extract_video(media, programme_id))
 361             elif kind == 'captions':
 362                 subtitles = self.extract_subtitles(media, programme_id)
 363         return formats, subtitles
 364
 365     def _download_playlist(self, playlist_id):
 366         try:
 367             playlist = self._download_json(
 368                 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
 369                 playlist_id, 'Downloading playlist JSON')
 370
 371             version = playlist.get('defaultAvailableVersion')
 372             if version:
 373                 smp_config = version['smpConfig']
 374                 title = smp_config['title']
 375                 description = smp_config['summary']
 376                 for item in smp_config['items']:
 377                     kind = item['kind']
 378                     if kind != 'programme' and kind != 'radioProgramme':
 379                         continue
 380                     programme_id = item.get('vpid')
 381                     duration = int_or_none(item.get('duration'))
 382                     formats, subtitles = self._download_media_selector(programme_id)
 383                 return programme_id, title, description, duration, formats, subtitles
 384         except ExtractorError as ee:
 385             if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
 386                 raise
 387
 388         # fallback to legacy playlist
 389         return self._process_legacy_playlist(playlist_id)
 390
 391     def _process_legacy_playlist_url(self, url, display_id):
 392         playlist = self._download_legacy_playlist_url(url, display_id)
 393         return self._extract_from_legacy_playlist(playlist, display_id)
 394
 395     def _process_legacy_playlist(self, playlist_id):
 396         return self._process_legacy_playlist_url(
 397             'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
 398
 399     def _download_legacy_playlist_url(self, url, playlist_id=None):
 400         return self._download_xml(
 401             url, playlist_id, 'Downloading legacy playlist XML')
 402
 403     def _extract_from_legacy_playlist(self, playlist, playlist_id):
 404         no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS)
 405         if no_items is not None:
 406             reason = no_items.get('reason')
 407             if reason == 'preAvailability':
 408                 msg = 'Episode %s is not yet available' % playlist_id
 409             elif reason == 'postAvailability':
 410                 msg = 'Episode %s is no longer available' % playlist_id
 411             elif reason == 'noMedia':
 412                 msg = 'Episode %s is not currently available' % playlist_id
 413             else:
 414                 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
 415             raise ExtractorError(msg, expected=True)
 416
 417         for item in self._extract_items(playlist):
 418             kind = item.get('kind')
 419             if kind != 'programme' and kind != 'radioProgramme':
 420                 continue
 421             title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
 422             description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
 423             description = description_el.text if description_el else None
 424
 425             def get_programme_id(item):
 426                 def get_from_attributes(item):
 427                     for p in('identifier', 'group'):
 428                         value = item.get(p)
 429                         if value and re.match(r'^[pb][\da-z]{7}$', value):
 430                             return value
 431                 get_from_attributes(item)
 432                 mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS)
 433                 if mediator is not None:
 434                     return get_from_attributes(mediator)
 435
 436             programme_id = get_programme_id(item)
 437             duration = int_or_none(item.get('duration'))
 438
 439             if programme_id:
 440                 formats, subtitles = self._download_media_selector(programme_id)
 441             else:
 442                 formats, subtitles = self._process_media_selector(item, playlist_id)
 443                 programme_id = playlist_id
 444
 445         return programme_id, title, description, duration, formats, subtitles
 446
 447     def _real_extract(self, url):
 448         group_id = self._match_id(url)
 449
 450         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 451
 452         programme_id = None
 453
 454         tviplayer = self._search_regex(
 455             r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
 456             webpage, 'player', default=None)
 457
 458         if tviplayer:
 459             player = self._parse_json(tviplayer, group_id).get('player', {})
 460             duration = int_or_none(player.get('duration'))
 461             programme_id = player.get('vpid')
 462
 463         if not programme_id:
 464             programme_id = self._search_regex(
 465                 r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
 466
 467         if programme_id:
 468             formats, subtitles = self._download_media_selector(programme_id)
 469             title = self._og_search_title(webpage)
 470             description = self._search_regex(
 471                 r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
 472                 webpage, 'description', fatal=False)
 473         else:
 474             programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
 475
 476         self._sort_formats(formats)
 477
 478         return {
 479             'id': programme_id,
 480             'title': title,
 481             'description': description,
 482             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 483             'duration': duration,
 484             'formats': formats,
 485             'subtitles': subtitles,
 486         }
 487
 488
 489 class BBCIE(BBCCoUkIE):
 490     IE_NAME = 'bbc'
 491     IE_DESC = 'BBC'
 492     _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
 493
 494     _MEDIASELECTOR_URLS = [
 495         # Provides more formats, namely direct mp4 links, but fails on some videos with
 496         # notukerror for non UK (?) users (e.g.
 497         # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
 498         'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
 499         # Provides fewer formats, but works everywhere for everybody (hopefully)
 500         'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
 501     ]
 502
 503     _TESTS = [{
 504         # article with multiple videos embedded with data-media-meta containing
 505         # playlist.sxml, externalId and no direct video links
 506         'url': 'http://www.bbc.com/news/world-europe-32668511',
 507         'info_dict': {
 508             'id': 'world-europe-32668511',
 509             'title': 'Russia stages massive WW2 parade despite Western boycott',
 510             'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
 511         },
 512         'playlist_count': 2,
 513     }, {
 514         # article with multiple videos embedded with data-media-meta (more videos)
 515         'url': 'http://www.bbc.com/news/business-28299555',
 516         'info_dict': {
 517             'id': 'business-28299555',
 518             'title': 'Farnborough Airshow: Video highlights',
 519             'description': 'BBC reports and video highlights at the Farnborough Airshow.',
 520         },
 521         'playlist_count': 9,
 522         'skip': 'Save time',
 523     }, {
 524         # article with multiple videos embedded with `new SMP()`
 525         'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
 526         'info_dict': {
 527             'id': '3662a707-0af9-3149-963f-47bea720b460',
 528             'title': 'BBC Blogs - Adam Curtis - BUGGER',
 529         },
 530         'playlist_count': 18,
 531     }, {
 532         # single video embedded with mediaAssetPage.init()
 533         'url': 'http://www.bbc.com/news/world-europe-32041533',
 534         'info_dict': {
 535             'id': 'p02mprgb',
 536             'ext': 'mp4',
 537             'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
 538             'duration': 47,
 539             'timestamp': 1427219242,
 540             'upload_date': '20150324',
 541         },
 542         'params': {
 543             # rtmp download
 544             'skip_download': True,
 545         }
 546     }, {
 547         # article with single video embedded with data-media-meta containing
 548         # direct video links (for now these are extracted) and playlist.xml (with
 549         # media items as f4m and m3u8 - currently unsupported)
 550         'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
 551         'info_dict': {
 552             'id': '150615_telabyad_kentin_cogu',
 553             'ext': 'mp4',
 554             'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
 555             'duration': 47,
 556             'timestamp': 1434397334,
 557             'upload_date': '20150615',
 558         },
 559         'params': {
 560             'skip_download': True,
 561         }
 562     }, {
 563         # single video embedded with playlist.sxml in data-playable
 564         'url': 'http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani',
 565         'info_dict': {
 566             'id': '151010_vid_ankara_patlama_ani',
 567             'ext': 'mp4',
 568             'title': "Ankara'da patlama anı",
 569             'timestamp': 1444480325,
 570             'upload_date': '20151010',
 571         },
 572         'params': {
 573             'skip_download': True,
 574         }
 575     }, {
 576         # single video embedded with mediaAssetPage.init() (regional section)
 577         'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
 578         'info_dict': {
 579             'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
 580             'ext': 'mp4',
 581             'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
 582             'duration': 87,
 583             'timestamp': 1434713142,
 584             'upload_date': '20150619',
 585         },
 586         'params': {
 587             'skip_download': True,
 588         }
 589     }, {
 590         # single video from video playlist embedded with vxp-playlist-data JSON
 591         'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
 592         'info_dict': {
 593             'id': 'p02w6qjc',
 594             'ext': 'mp4',
 595             'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
 596             'duration': 56,
 597         },
 598         'params': {
 599             'skip_download': True,
 600         }
 601     }, {
 602         # single video story with digitalData
 603         'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
 604         'info_dict': {
 605             'id': 'p02q6gc4',
 606             'ext': 'flv',
 607             'title': 'Sri Lanka’s spicy secret',
 608             'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
 609             'timestamp': 1437674293,
 610             'upload_date': '20150723',
 611         },
 612         'params': {
 613             # rtmp download
 614             'skip_download': True,
 615         }
 616     }, {
 617         # single video story without digitalData
 618         'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
 619         'info_dict': {
 620             'id': 'p018zqqg',
 621             'ext': 'mp4',
 622             'title': 'Hyundai Santa Fe Sport: Rock star',
 623             'description': 'md5:b042a26142c4154a6e472933cf20793d',
 624             'timestamp': 1368473503,
 625             'upload_date': '20130513',
 626         },
 627         'params': {
 628             # rtmp download
 629             'skip_download': True,
 630         }
 631     }, {
 632         # single video with playlist.sxml URL
 633         'url': 'http://www.bbc.com/sport/0/football/33653409',
 634         'info_dict': {
 635             'id': 'p02xycnp',
 636             'ext': 'mp4',
 637             'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
 638             'description': 'md5:398fca0e2e701c609d726e034fa1fc89',
 639             'duration': 140,
 640         },
 641         'params': {
 642             # rtmp download
 643             'skip_download': True,
 644         }
 645     }, {
 646         # article with multiple videos embedded with playlist.sxml
 647         'url': 'http://www.bbc.com/sport/0/football/34475836',
 648         'info_dict': {
 649             'id': '34475836',
 650             'title': 'What Liverpool can expect from Klopp',
 651         },
 652         'playlist_count': 3,
 653     }, {
 654         # single video with playlist URL from weather section
 655         'url': 'http://www.bbc.com/weather/features/33601775',
 656         'only_matching': True,
 657     }, {
 658         # custom redirection to www.bbc.com
 659         'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
 660         'only_matching': True,
 661     }]
 662
 663     @classmethod
 664     def suitable(cls, url):
 665         return False if BBCCoUkIE.suitable(url) else super(BBCIE, cls).suitable(url)
 666
 667     def _extract_from_media_meta(self, media_meta, video_id):
 668         # Direct links to media in media metadata (e.g.
 669         # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
 670         # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
 671         source_files = media_meta.get('sourceFiles')
 672         if source_files:
 673             return [{
 674                 'url': f['url'],
 675                 'format_id': format_id,
 676                 'ext': f.get('encoding'),
 677                 'tbr': float_or_none(f.get('bitrate'), 1000),
 678                 'filesize': int_or_none(f.get('filesize')),
 679             } for format_id, f in source_files.items() if f.get('url')], []
 680
 681         programme_id = media_meta.get('externalId')
 682         if programme_id:
 683             return self._download_media_selector(programme_id)
 684
 685         # Process playlist.sxml as legacy playlist
 686         href = media_meta.get('href')
 687         if href:
 688             playlist = self._download_legacy_playlist_url(href)
 689             _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
 690             return formats, subtitles
 691
 692         return [], []
 693
 694     def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
 695         programme_id, title, description, duration, formats, subtitles = \
 696             self._process_legacy_playlist_url(url, playlist_id)
 697         self._sort_formats(formats)
 698         return {
 699             'id': programme_id,
 700             'title': title,
 701             'description': description,
 702             'duration': duration,
 703             'timestamp': timestamp,
 704             'formats': formats,
 705             'subtitles': subtitles,
 706         }
 707
 708     def _real_extract(self, url):
 709         playlist_id = self._match_id(url)
 710
 711         webpage = self._download_webpage(url, playlist_id)
 712
 713         timestamp = parse_iso8601(self._search_regex(
 714             [r'"datePublished":\s*"([^"]+)',
 715              r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
 716              r'itemprop="datePublished"[^>]+datetime="([^"]+)"'],
 717             webpage, 'date', default=None))
 718
 719         entries = []
 720
 721         # article with multiple videos embedded with playlist.sxml (e.g.
 722         # http://www.bbc.com/sport/0/football/34475836)
 723         playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
 724         if playlists:
 725             entries = [
 726                 self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
 727                 for playlist_url in playlists]
 728
 729         # news article with multiple videos embedded with data-playable
 730         data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
 731         if data_playables:
 732             for _, data_playable_json in data_playables:
 733                 data_playable = self._parse_json(
 734                     unescapeHTML(data_playable_json), playlist_id, fatal=False)
 735                 if not data_playable:
 736                     continue
 737                 settings = data_playable.get('settings', {})
 738                 if settings:
 739                     # data-playable with video vpid in settings.playlistObject.items (e.g.
 740                     # http://www.bbc.com/news/world-us-canada-34473351)
 741                     playlist_object = settings.get('playlistObject', {})
 742                     if playlist_object:
 743                         items = playlist_object.get('items')
 744                         if items and isinstance(items, list):
 745                             title = playlist_object['title']
 746                             description = playlist_object.get('summary')
 747                             duration = int_or_none(items[0].get('duration'))
 748                             programme_id = items[0].get('vpid')
 749                             formats, subtitles = self._download_media_selector(programme_id)
 750                             self._sort_formats(formats)
 751                             entries.append({
 752                                 'id': programme_id,
 753                                 'title': title,
 754                                 'description': description,
 755                                 'timestamp': timestamp,
 756                                 'duration': duration,
 757                                 'formats': formats,
 758                                 'subtitles': subtitles,
 759                             })
 760                     else:
 761                         # data-playable without vpid but with a playlist.sxml URLs
 762                         # in otherSettings.playlist (e.g.
 763                         # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
 764                         playlist = data_playable.get('otherSettings', {}).get('playlist', {})
 765                         if playlist:
 766                             entries.append(self._extract_from_playlist_sxml(
 767                                 playlist.get('progressiveDownloadUrl'), playlist_id, timestamp))
 768
 769         if entries:
 770             playlist_title = self._og_search_title(webpage)
 771             playlist_description = self._og_search_description(webpage, default=None)
 772             return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 773
 774         # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
 775         programme_id = self._search_regex(
 776             [r'data-video-player-vpid="([\da-z]{8})"',
 777              r'<param[^>]+name="externalIdentifier"[^>]+value="([\da-z]{8})"'],
 778             webpage, 'vpid', default=None)
 779
 780         if programme_id:
 781             formats, subtitles = self._download_media_selector(programme_id)
 782             self._sort_formats(formats)
 783             # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
 784             digital_data = self._parse_json(
 785                 self._search_regex(
 786                     r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
 787                 programme_id, fatal=False)
 788             page_info = digital_data.get('page', {}).get('pageInfo', {})
 789             title = page_info.get('pageName') or self._og_search_title(webpage)
 790             description = page_info.get('description') or self._og_search_description(webpage)
 791             timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
 792             return {
 793                 'id': programme_id,
 794                 'title': title,
 795                 'description': description,
 796                 'timestamp': timestamp,
 797                 'duration': duration,
 798                 'formats': formats,
 799                 'subtitles': subtitles,
 800             }
 801
 802         playlist_title = self._html_search_regex(
 803             r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'playlist title')
 804         playlist_description = self._og_search_description(webpage, default=None)
 805
 806         def extract_all(pattern):
 807             return list(filter(None, map(
 808                 lambda s: self._parse_json(s, playlist_id, fatal=False),
 809                 re.findall(pattern, webpage))))
 810
 811         # Multiple video article (e.g.
 812         # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
 813         EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+[\da-z]{8}(?:\b[^"]+)?'
 814         entries = []
 815         for match in extract_all(r'new\s+SMP\(({.+?})\)'):
 816             embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
 817             if embed_url and re.match(EMBED_URL, embed_url):
 818                 entries.append(embed_url)
 819         entries.extend(re.findall(
 820             r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
 821         if entries:
 822             return self.playlist_result(
 823                 [self.url_result(entry, 'BBCCoUk') for entry in entries],
 824                 playlist_id, playlist_title, playlist_description)
 825
 826         # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
 827         medias = extract_all(r"data-media-meta='({[^']+})'")
 828
 829         if not medias:
 830             # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
 831             media_asset = self._search_regex(
 832                 r'mediaAssetPage\.init\(\s*({.+?}), "/',
 833                 webpage, 'media asset', default=None)
 834             if media_asset:
 835                 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
 836                 medias = []
 837                 for video in media_asset_page.get('videos', {}).values():
 838                     medias.extend(video.values())
 839
 840         if not medias:
 841             # Multiple video playlist with single `now playing` entry (e.g.
 842             # http://www.bbc.com/news/video_and_audio/must_see/33767813)
 843             vxp_playlist = self._parse_json(
 844                 self._search_regex(
 845                     r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>',
 846                     webpage, 'playlist data'),
 847                 playlist_id)
 848             playlist_medias = []
 849             for item in vxp_playlist:
 850                 media = item.get('media')
 851                 if not media:
 852                     continue
 853                 playlist_medias.append(media)
 854                 # Download single video if found media with asset id matching the video id from URL
 855                 if item.get('advert', {}).get('assetId') == playlist_id:
 856                     medias = [media]
 857                     break
 858             # Fallback to the whole playlist
 859             if not medias:
 860                 medias = playlist_medias
 861
 862         entries = []
 863         for num, media_meta in enumerate(medias, start=1):
 864             formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
 865             if not formats:
 866                 continue
 867             self._sort_formats(formats)
 868
 869             video_id = media_meta.get('externalId')
 870             if not video_id:
 871                 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
 872
 873             title = media_meta.get('caption')
 874             if not title:
 875                 title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
 876
 877             duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
 878
 879             images = []
 880             for image in media_meta.get('images', {}).values():
 881                 images.extend(image.values())
 882             if 'image' in media_meta:
 883                 images.append(media_meta['image'])
 884
 885             thumbnails = [{
 886                 'url': image.get('href'),
 887                 'width': int_or_none(image.get('width')),
 888                 'height': int_or_none(image.get('height')),
 889             } for image in images]
 890
 891             entries.append({
 892                 'id': video_id,
 893                 'title': title,
 894                 'thumbnails': thumbnails,
 895                 'duration': duration,
 896                 'timestamp': timestamp,
 897                 'formats': formats,
 898                 'subtitles': subtitles,
 899             })
 900
 901         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)