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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import itertools
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     clean_html,
  10     dict_get,
  11     ExtractorError,
  12     float_or_none,
  13     get_element_by_class,
  14     int_or_none,
  15     js_to_json,
  16     parse_duration,
  17     parse_iso8601,
  18     try_get,
  19     unescapeHTML,
  20     urlencode_postdata,
  21     urljoin,
  22 )
  23 from ..compat import (
  24     compat_HTTPError,
  25     compat_urlparse,
  26 )
  27
  28
  29 class BBCCoUkIE(InfoExtractor):
  30     IE_NAME = 'bbc.co.uk'
  31     IE_DESC = 'BBC iPlayer'
  32     _ID_REGEX = r'[pbw][\da-z]{7}'
  33     _VALID_URL = r'''(?x)
  34                     https?://
  35                         (?:www\.)?bbc\.co\.uk/
  36                         (?:
  37                             programmes/(?!articles/)|
  38                             iplayer(?:/[^/]+)?/(?:episode/|playlist/)|
  39                             music/(?:clips|audiovideo/popular)[/#]|
  40                             radio/player/|
  41                             events/[^/]+/play/[^/]+/
  42                         )
  43                         (?P<id>%s)(?!/(?:episodes|broadcasts|clips))
  44                     ''' % _ID_REGEX
  45
  46     _LOGIN_URL = 'https://account.bbc.com/signin'
  47     _NETRC_MACHINE = 'bbc'
  48
  49     _MEDIASELECTOR_URLS = [
  50         # Provides HQ HLS streams with even better quality that pc mediaset but fails
  51         # with geolocation in some cases when it's even not geo restricted at all (e.g.
  52         # http://www.bbc.co.uk/programmes/b06bp7lf). Also may fail with selectionunavailable.
  53         'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
  54         'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
  55     ]
  56
  57     _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
  58     _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
  59
  60     _NAMESPACES = (
  61         _MEDIASELECTION_NS,
  62         _EMP_PLAYLIST_NS,
  63     )
  64
  65     _TESTS = [
  66         {
  67             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
  68             'info_dict': {
  69                 'id': 'b039d07m',
  70                 'ext': 'flv',
  71                 'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
  72                 'description': 'The Canadian poet and songwriter reflects on his musical career.',
  73             },
  74             'params': {
  75                 # rtmp download
  76                 'skip_download': True,
  77             }
  78         },
  79         {
  80             'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
  81             'info_dict': {
  82                 'id': 'b00yng1d',
  83                 'ext': 'flv',
  84                 'title': 'The Man in Black: Series 3: The Printed Name',
  85                 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
  86                 'duration': 1800,
  87             },
  88             'params': {
  89                 # rtmp download
  90                 'skip_download': True,
  91             },
  92             'skip': 'Episode is no longer available on BBC iPlayer Radio',
  93         },
  94         {
  95             'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
  96             'info_dict': {
  97                 'id': 'b00yng1d',
  98                 'ext': 'flv',
  99                 'title': 'The Voice UK: Series 3: Blind Auditions 5',
 100                 '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.',
 101                 'duration': 5100,
 102             },
 103             'params': {
 104                 # rtmp download
 105                 'skip_download': True,
 106             },
 107             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
 108         },
 109         {
 110             'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
 111             'info_dict': {
 112                 'id': 'b03k3pb7',
 113                 'ext': 'flv',
 114                 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
 115                 'description': '2. Invasion',
 116                 'duration': 3600,
 117             },
 118             'params': {
 119                 # rtmp download
 120                 'skip_download': True,
 121             },
 122             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
 123         }, {
 124             'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
 125             'info_dict': {
 126                 'id': 'b04v209v',
 127                 'ext': 'flv',
 128                 'title': 'Pete Tong, The Essential New Tune Special',
 129                 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
 130                 'duration': 10800,
 131             },
 132             'params': {
 133                 # rtmp download
 134                 'skip_download': True,
 135             },
 136             'skip': 'Episode is no longer available on BBC iPlayer Radio',
 137         }, {
 138             'url': 'http://www.bbc.co.uk/music/clips/p022h44b',
 139             'note': 'Audio',
 140             'info_dict': {
 141                 'id': 'p022h44j',
 142                 'ext': 'flv',
 143                 'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
 144                 'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
 145                 'duration': 227,
 146             },
 147             'params': {
 148                 # rtmp download
 149                 'skip_download': True,
 150             }
 151         }, {
 152             'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
 153             'note': 'Video',
 154             'info_dict': {
 155                 'id': 'p025c103',
 156                 'ext': 'flv',
 157                 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
 158                 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
 159                 'duration': 226,
 160             },
 161             'params': {
 162                 # rtmp download
 163                 'skip_download': True,
 164             }
 165         }, {
 166             'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
 167             'info_dict': {
 168                 'id': 'p02n76xf',
 169                 'ext': 'flv',
 170                 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
 171                 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
 172                 'duration': 3540,
 173             },
 174             'params': {
 175                 # rtmp download
 176                 'skip_download': True,
 177             },
 178             'skip': 'geolocation',
 179         }, {
 180             'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
 181             'info_dict': {
 182                 'id': 'b05zmgw1',
 183                 'ext': 'flv',
 184                 '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.',
 185                 'title': 'Royal Academy Summer Exhibition',
 186                 'duration': 3540,
 187             },
 188             'params': {
 189                 # rtmp download
 190                 'skip_download': True,
 191             },
 192             'skip': 'geolocation',
 193         }, {
 194             # iptv-all mediaset fails with geolocation however there is no geo restriction
 195             # for this programme at all
 196             'url': 'http://www.bbc.co.uk/programmes/b06rkn85',
 197             'info_dict': {
 198                 'id': 'b06rkms3',
 199                 'ext': 'flv',
 200                 'title': "Best of the Mini-Mixes 2015: Part 3, Annie Mac's Friday Night - BBC Radio 1",
 201                 'description': "Annie has part three in the Best of the Mini-Mixes 2015, plus the year's Most Played!",
 202             },
 203             'params': {
 204                 # rtmp download
 205                 'skip_download': True,
 206             },
 207             'skip': 'Now it\'s really geo-restricted',
 208         }, {
 209             # compact player (https://github.com/rg3/youtube-dl/issues/8147)
 210             'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
 211             'info_dict': {
 212                 'id': 'p028bfkj',
 213                 'ext': 'flv',
 214                 'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
 215                 'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
 216             },
 217             'params': {
 218                 # rtmp download
 219                 'skip_download': True,
 220             },
 221         }, {
 222             'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
 223             'only_matching': True,
 224         }, {
 225             'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
 226             'only_matching': True,
 227         }, {
 228             'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
 229             'only_matching': True,
 230         }, {
 231             'url': 'http://www.bbc.co.uk/radio/player/p03cchwf',
 232             'only_matching': True,
 233         }, {
 234             'url': 'https://www.bbc.co.uk/music/audiovideo/popular#p055bc55',
 235             'only_matching': True,
 236         }, {
 237             'url': 'http://www.bbc.co.uk/programmes/w3csv1y9',
 238             'only_matching': True,
 239         }]
 240
 241     _USP_RE = r'/([^/]+?)\.ism(?:\.hlsv2\.ism)?/[^/]+\.m3u8'
 242
 243     def _login(self):
 244         username, password = self._get_login_info()
 245         if username is None:
 246             return
 247
 248         login_page = self._download_webpage(
 249             self._LOGIN_URL, None, 'Downloading signin page')
 250
 251         login_form = self._hidden_inputs(login_page)
 252
 253         login_form.update({
 254             'username': username,
 255             'password': password,
 256         })
 257
 258         post_url = urljoin(self._LOGIN_URL, self._search_regex(
 259             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
 260             'post url', default=self._LOGIN_URL, group='url'))
 261
 262         response, urlh = self._download_webpage_handle(
 263             post_url, None, 'Logging in', data=urlencode_postdata(login_form),
 264             headers={'Referer': self._LOGIN_URL})
 265
 266         if self._LOGIN_URL in urlh.geturl():
 267             error = clean_html(get_element_by_class('form-message', response))
 268             if error:
 269                 raise ExtractorError(
 270                     'Unable to login: %s' % error, expected=True)
 271             raise ExtractorError('Unable to log in')
 272
 273     def _real_initialize(self):
 274         self._login()
 275
 276     class MediaSelectionError(Exception):
 277         def __init__(self, id):
 278             self.id = id
 279
 280     def _extract_asx_playlist(self, connection, programme_id):
 281         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
 282         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
 283
 284     def _extract_items(self, playlist):
 285         return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
 286
 287     def _findall_ns(self, element, xpath):
 288         elements = []
 289         for ns in self._NAMESPACES:
 290             elements.extend(element.findall(xpath % ns))
 291         return elements
 292
 293     def _extract_medias(self, media_selection):
 294         error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS)
 295         if error is None:
 296             media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS)
 297         if error is not None:
 298             raise BBCCoUkIE.MediaSelectionError(error.get('id'))
 299         return self._findall_ns(media_selection, './{%s}media')
 300
 301     def _extract_connections(self, media):
 302         return self._findall_ns(media, './{%s}connection')
 303
 304     def _get_subtitles(self, media, programme_id):
 305         subtitles = {}
 306         for connection in self._extract_connections(media):
 307             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 308             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 309             subtitles[lang] = [
 310                 {
 311                     'url': connection.get('href'),
 312                     'ext': 'ttml',
 313                 },
 314             ]
 315         return subtitles
 316
 317     def _raise_extractor_error(self, media_selection_error):
 318         raise ExtractorError(
 319             '%s returned error: %s' % (self.IE_NAME, media_selection_error.id),
 320             expected=True)
 321
 322     def _download_media_selector(self, programme_id):
 323         last_exception = None
 324         for mediaselector_url in self._MEDIASELECTOR_URLS:
 325             try:
 326                 return self._download_media_selector_url(
 327                     mediaselector_url % programme_id, programme_id)
 328             except BBCCoUkIE.MediaSelectionError as e:
 329                 if e.id in ('notukerror', 'geolocation', 'selectionunavailable'):
 330                     last_exception = e
 331                     continue
 332                 self._raise_extractor_error(e)
 333         self._raise_extractor_error(last_exception)
 334
 335     def _download_media_selector_url(self, url, programme_id=None):
 336         media_selection = self._download_xml(
 337             url, programme_id, 'Downloading media selection XML',
 338             expected_status=(403, 404))
 339         return self._process_media_selector(media_selection, programme_id)
 340
 341     def _process_media_selector(self, media_selection, programme_id):
 342         formats = []
 343         subtitles = None
 344         urls = []
 345
 346         for media in self._extract_medias(media_selection):
 347             kind = media.get('kind')
 348             if kind in ('video', 'audio'):
 349                 bitrate = int_or_none(media.get('bitrate'))
 350                 encoding = media.get('encoding')
 351                 service = media.get('service')
 352                 width = int_or_none(media.get('width'))
 353                 height = int_or_none(media.get('height'))
 354                 file_size = int_or_none(media.get('media_file_size'))
 355                 for connection in self._extract_connections(media):
 356                     href = connection.get('href')
 357                     if href in urls:
 358                         continue
 359                     if href:
 360                         urls.append(href)
 361                     conn_kind = connection.get('kind')
 362                     protocol = connection.get('protocol')
 363                     supplier = connection.get('supplier')
 364                     transfer_format = connection.get('transferFormat')
 365                     format_id = supplier or conn_kind or protocol
 366                     if service:
 367                         format_id = '%s_%s' % (service, format_id)
 368                     # ASX playlist
 369                     if supplier == 'asx':
 370                         for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
 371                             formats.append({
 372                                 'url': ref,
 373                                 'format_id': 'ref%s_%s' % (i, format_id),
 374                             })
 375                     elif transfer_format == 'dash':
 376                         formats.extend(self._extract_mpd_formats(
 377                             href, programme_id, mpd_id=format_id, fatal=False))
 378                     elif transfer_format == 'hls':
 379                         formats.extend(self._extract_m3u8_formats(
 380                             href, programme_id, ext='mp4', entry_protocol='m3u8_native',
 381                             m3u8_id=format_id, fatal=False))
 382                         if re.search(self._USP_RE, href):
 383                             usp_formats = self._extract_m3u8_formats(
 384                                 re.sub(self._USP_RE, r'/\1.ism/\1.m3u8', href),
 385                                 programme_id, ext='mp4', entry_protocol='m3u8_native',
 386                                 m3u8_id=format_id, fatal=False)
 387                             for f in usp_formats:
 388                                 if f.get('height') and f['height'] > 720:
 389                                     continue
 390                                 formats.append(f)
 391                     elif transfer_format == 'hds':
 392                         formats.extend(self._extract_f4m_formats(
 393                             href, programme_id, f4m_id=format_id, fatal=False))
 394                     else:
 395                         if not service and not supplier and bitrate:
 396                             format_id += '-%d' % bitrate
 397                         fmt = {
 398                             'format_id': format_id,
 399                             'filesize': file_size,
 400                         }
 401                         if kind == 'video':
 402                             fmt.update({
 403                                 'width': width,
 404                                 'height': height,
 405                                 'tbr': bitrate,
 406                                 'vcodec': encoding,
 407                             })
 408                         else:
 409                             fmt.update({
 410                                 'abr': bitrate,
 411                                 'acodec': encoding,
 412                                 'vcodec': 'none',
 413                             })
 414                         if protocol in ('http', 'https'):
 415                             # Direct link
 416                             fmt.update({
 417                                 'url': href,
 418                             })
 419                         elif protocol == 'rtmp':
 420                             application = connection.get('application', 'ondemand')
 421                             auth_string = connection.get('authString')
 422                             identifier = connection.get('identifier')
 423                             server = connection.get('server')
 424                             fmt.update({
 425                                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 426                                 'play_path': identifier,
 427                                 'app': '%s?%s' % (application, auth_string),
 428                                 'page_url': 'http://www.bbc.co.uk',
 429                                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 430                                 'rtmp_live': False,
 431                                 'ext': 'flv',
 432                             })
 433                         else:
 434                             continue
 435                         formats.append(fmt)
 436             elif kind == 'captions':
 437                 subtitles = self.extract_subtitles(media, programme_id)
 438         return formats, subtitles
 439
 440     def _download_playlist(self, playlist_id):
 441         try:
 442             playlist = self._download_json(
 443                 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
 444                 playlist_id, 'Downloading playlist JSON')
 445
 446             version = playlist.get('defaultAvailableVersion')
 447             if version:
 448                 smp_config = version['smpConfig']
 449                 title = smp_config['title']
 450                 description = smp_config['summary']
 451                 for item in smp_config['items']:
 452                     kind = item['kind']
 453                     if kind not in ('programme', 'radioProgramme'):
 454                         continue
 455                     programme_id = item.get('vpid')
 456                     duration = int_or_none(item.get('duration'))
 457                     formats, subtitles = self._download_media_selector(programme_id)
 458                 return programme_id, title, description, duration, formats, subtitles
 459         except ExtractorError as ee:
 460             if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
 461                 raise
 462
 463         # fallback to legacy playlist
 464         return self._process_legacy_playlist(playlist_id)
 465
 466     def _process_legacy_playlist_url(self, url, display_id):
 467         playlist = self._download_legacy_playlist_url(url, display_id)
 468         return self._extract_from_legacy_playlist(playlist, display_id)
 469
 470     def _process_legacy_playlist(self, playlist_id):
 471         return self._process_legacy_playlist_url(
 472             'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
 473
 474     def _download_legacy_playlist_url(self, url, playlist_id=None):
 475         return self._download_xml(
 476             url, playlist_id, 'Downloading legacy playlist XML')
 477
 478     def _extract_from_legacy_playlist(self, playlist, playlist_id):
 479         no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS)
 480         if no_items is not None:
 481             reason = no_items.get('reason')
 482             if reason == 'preAvailability':
 483                 msg = 'Episode %s is not yet available' % playlist_id
 484             elif reason == 'postAvailability':
 485                 msg = 'Episode %s is no longer available' % playlist_id
 486             elif reason == 'noMedia':
 487                 msg = 'Episode %s is not currently available' % playlist_id
 488             else:
 489                 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
 490             raise ExtractorError(msg, expected=True)
 491
 492         for item in self._extract_items(playlist):
 493             kind = item.get('kind')
 494             if kind not in ('programme', 'radioProgramme'):
 495                 continue
 496             title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
 497             description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
 498             description = description_el.text if description_el is not None else None
 499
 500             def get_programme_id(item):
 501                 def get_from_attributes(item):
 502                     for p in('identifier', 'group'):
 503                         value = item.get(p)
 504                         if value and re.match(r'^[pb][\da-z]{7}$', value):
 505                             return value
 506                 get_from_attributes(item)
 507                 mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS)
 508                 if mediator is not None:
 509                     return get_from_attributes(mediator)
 510
 511             programme_id = get_programme_id(item)
 512             duration = int_or_none(item.get('duration'))
 513
 514             if programme_id:
 515                 formats, subtitles = self._download_media_selector(programme_id)
 516             else:
 517                 formats, subtitles = self._process_media_selector(item, playlist_id)
 518                 programme_id = playlist_id
 519
 520         return programme_id, title, description, duration, formats, subtitles
 521
 522     def _real_extract(self, url):
 523         group_id = self._match_id(url)
 524
 525         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 526
 527         error = self._search_regex(
 528             r'<div\b[^>]+\bclass=["\']smp__message delta["\'][^>]*>([^<]+)<',
 529             webpage, 'error', default=None)
 530         if error:
 531             raise ExtractorError(error, expected=True)
 532
 533         programme_id = None
 534         duration = None
 535
 536         tviplayer = self._search_regex(
 537             r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
 538             webpage, 'player', default=None)
 539
 540         if tviplayer:
 541             player = self._parse_json(tviplayer, group_id).get('player', {})
 542             duration = int_or_none(player.get('duration'))
 543             programme_id = player.get('vpid')
 544
 545         if not programme_id:
 546             programme_id = self._search_regex(
 547                 r'"vpid"\s*:\s*"(%s)"' % self._ID_REGEX, webpage, 'vpid', fatal=False, default=None)
 548
 549         if programme_id:
 550             formats, subtitles = self._download_media_selector(programme_id)
 551             title = self._og_search_title(webpage, default=None) or self._html_search_regex(
 552                 (r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>',
 553                  r'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage, 'title')
 554             description = self._search_regex(
 555                 (r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
 556                  r'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'),
 557                 webpage, 'description', default=None)
 558             if not description:
 559                 description = self._html_search_meta('description', webpage)
 560         else:
 561             programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
 562
 563         self._sort_formats(formats)
 564
 565         return {
 566             'id': programme_id,
 567             'title': title,
 568             'description': description,
 569             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 570             'duration': duration,
 571             'formats': formats,
 572             'subtitles': subtitles,
 573         }
 574
 575
 576 class BBCIE(BBCCoUkIE):
 577     IE_NAME = 'bbc'
 578     IE_DESC = 'BBC'
 579     _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
 580
 581     _MEDIASELECTOR_URLS = [
 582         # Provides HQ HLS streams but fails with geolocation in some cases when it's
 583         # even not geo restricted at all
 584         'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
 585         # Provides more formats, namely direct mp4 links, but fails on some videos with
 586         # notukerror for non UK (?) users (e.g.
 587         # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
 588         'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
 589         # Provides fewer formats, but works everywhere for everybody (hopefully)
 590         'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
 591     ]
 592
 593     _TESTS = [{
 594         # article with multiple videos embedded with data-playable containing vpids
 595         'url': 'http://www.bbc.com/news/world-europe-32668511',
 596         'info_dict': {
 597             'id': 'world-europe-32668511',
 598             'title': 'Russia stages massive WW2 parade despite Western boycott',
 599             'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
 600         },
 601         'playlist_count': 2,
 602     }, {
 603         # article with multiple videos embedded with data-playable (more videos)
 604         'url': 'http://www.bbc.com/news/business-28299555',
 605         'info_dict': {
 606             'id': 'business-28299555',
 607             'title': 'Farnborough Airshow: Video highlights',
 608             'description': 'BBC reports and video highlights at the Farnborough Airshow.',
 609         },
 610         'playlist_count': 9,
 611         'skip': 'Save time',
 612     }, {
 613         # article with multiple videos embedded with `new SMP()`
 614         # broken
 615         'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
 616         'info_dict': {
 617             'id': '3662a707-0af9-3149-963f-47bea720b460',
 618             'title': 'BUGGER',
 619         },
 620         'playlist_count': 18,
 621     }, {
 622         # single video embedded with data-playable containing vpid
 623         'url': 'http://www.bbc.com/news/world-europe-32041533',
 624         'info_dict': {
 625             'id': 'p02mprgb',
 626             'ext': 'mp4',
 627             'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
 628             'description': 'md5:2868290467291b37feda7863f7a83f54',
 629             'duration': 47,
 630             'timestamp': 1427219242,
 631             'upload_date': '20150324',
 632         },
 633         'params': {
 634             # rtmp download
 635             'skip_download': True,
 636         }
 637     }, {
 638         # article with single video embedded with data-playable containing XML playlist
 639         # with direct video links as progressiveDownloadUrl (for now these are extracted)
 640         # and playlist with f4m and m3u8 as streamingUrl
 641         'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
 642         'info_dict': {
 643             'id': '150615_telabyad_kentin_cogu',
 644             'ext': 'mp4',
 645             'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
 646             'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
 647             'timestamp': 1434397334,
 648             'upload_date': '20150615',
 649         },
 650         'params': {
 651             'skip_download': True,
 652         }
 653     }, {
 654         # single video embedded with data-playable containing XML playlists (regional section)
 655         'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
 656         'info_dict': {
 657             'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
 658             'ext': 'mp4',
 659             'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
 660             'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
 661             'timestamp': 1434713142,
 662             'upload_date': '20150619',
 663         },
 664         'params': {
 665             'skip_download': True,
 666         }
 667     }, {
 668         # single video from video playlist embedded with vxp-playlist-data JSON
 669         'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
 670         'info_dict': {
 671             'id': 'p02w6qjc',
 672             'ext': 'mp4',
 673             'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
 674             'duration': 56,
 675             'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
 676         },
 677         'params': {
 678             'skip_download': True,
 679         }
 680     }, {
 681         # single video story with digitalData
 682         'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
 683         'info_dict': {
 684             'id': 'p02q6gc4',
 685             'ext': 'flv',
 686             'title': 'Sri Lanka’s spicy secret',
 687             'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
 688             'timestamp': 1437674293,
 689             'upload_date': '20150723',
 690         },
 691         'params': {
 692             # rtmp download
 693             'skip_download': True,
 694         }
 695     }, {
 696         # single video story without digitalData
 697         'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
 698         'info_dict': {
 699             'id': 'p018zqqg',
 700             'ext': 'mp4',
 701             'title': 'Hyundai Santa Fe Sport: Rock star',
 702             'description': 'md5:b042a26142c4154a6e472933cf20793d',
 703             'timestamp': 1415867444,
 704             'upload_date': '20141113',
 705         },
 706         'params': {
 707             # rtmp download
 708             'skip_download': True,
 709         }
 710     }, {
 711         # single video embedded with Morph
 712         'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
 713         'info_dict': {
 714             'id': 'p041vhd0',
 715             'ext': 'mp4',
 716             'title': "Nigeria v Japan - Men's First Round",
 717             'description': 'Live coverage of the first round from Group B at the Amazonia Arena.',
 718             'duration': 7980,
 719             'uploader': 'BBC Sport',
 720             'uploader_id': 'bbc_sport',
 721         },
 722         'params': {
 723             # m3u8 download
 724             'skip_download': True,
 725         },
 726         'skip': 'Georestricted to UK',
 727     }, {
 728         # single video with playlist.sxml URL in playlist param
 729         'url': 'http://www.bbc.com/sport/0/football/33653409',
 730         'info_dict': {
 731             'id': 'p02xycnp',
 732             'ext': 'mp4',
 733             'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
 734             'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
 735             'duration': 140,
 736         },
 737         'params': {
 738             # rtmp download
 739             'skip_download': True,
 740         }
 741     }, {
 742         # article with multiple videos embedded with playlist.sxml in playlist param
 743         'url': 'http://www.bbc.com/sport/0/football/34475836',
 744         'info_dict': {
 745             'id': '34475836',
 746             'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
 747             'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
 748         },
 749         'playlist_count': 3,
 750     }, {
 751         # school report article with single video
 752         'url': 'http://www.bbc.co.uk/schoolreport/35744779',
 753         'info_dict': {
 754             'id': '35744779',
 755             'title': 'School which breaks down barriers in Jerusalem',
 756         },
 757         'playlist_count': 1,
 758     }, {
 759         # single video with playlist URL from weather section
 760         'url': 'http://www.bbc.com/weather/features/33601775',
 761         'only_matching': True,
 762     }, {
 763         # custom redirection to www.bbc.com
 764         'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
 765         'only_matching': True,
 766     }, {
 767         # single video article embedded with data-media-vpid
 768         'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
 769         'only_matching': True,
 770     }, {
 771         'url': 'https://www.bbc.co.uk/bbcthree/clip/73d0bbd0-abc3-4cea-b3c0-cdae21905eb1',
 772         'info_dict': {
 773             'id': 'p06556y7',
 774             'ext': 'mp4',
 775             'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
 776             'description': 'md5:4b7dfd063d5a789a1512e99662be3ddd',
 777         },
 778         'params': {
 779             'skip_download': True,
 780         }
 781     }]
 782
 783     @classmethod
 784     def suitable(cls, url):
 785         EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE)
 786         return (False if any(ie.suitable(url) for ie in EXCLUDE_IE)
 787                 else super(BBCIE, cls).suitable(url))
 788
 789     def _extract_from_media_meta(self, media_meta, video_id):
 790         # Direct links to media in media metadata (e.g.
 791         # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
 792         # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
 793         source_files = media_meta.get('sourceFiles')
 794         if source_files:
 795             return [{
 796                 'url': f['url'],
 797                 'format_id': format_id,
 798                 'ext': f.get('encoding'),
 799                 'tbr': float_or_none(f.get('bitrate'), 1000),
 800                 'filesize': int_or_none(f.get('filesize')),
 801             } for format_id, f in source_files.items() if f.get('url')], []
 802
 803         programme_id = media_meta.get('externalId')
 804         if programme_id:
 805             return self._download_media_selector(programme_id)
 806
 807         # Process playlist.sxml as legacy playlist
 808         href = media_meta.get('href')
 809         if href:
 810             playlist = self._download_legacy_playlist_url(href)
 811             _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
 812             return formats, subtitles
 813
 814         return [], []
 815
 816     def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
 817         programme_id, title, description, duration, formats, subtitles = \
 818             self._process_legacy_playlist_url(url, playlist_id)
 819         self._sort_formats(formats)
 820         return {
 821             'id': programme_id,
 822             'title': title,
 823             'description': description,
 824             'duration': duration,
 825             'timestamp': timestamp,
 826             'formats': formats,
 827             'subtitles': subtitles,
 828         }
 829
 830     def _real_extract(self, url):
 831         playlist_id = self._match_id(url)
 832
 833         webpage = self._download_webpage(url, playlist_id)
 834
 835         json_ld_info = self._search_json_ld(webpage, playlist_id, default={})
 836         timestamp = json_ld_info.get('timestamp')
 837
 838         playlist_title = json_ld_info.get('title')
 839         if not playlist_title:
 840             playlist_title = self._og_search_title(
 841                 webpage, default=None) or self._html_search_regex(
 842                 r'<title>(.+?)</title>', webpage, 'playlist title', default=None)
 843             if playlist_title:
 844                 playlist_title = re.sub(r'(.+)\s*-\s*BBC.*?$', r'\1', playlist_title).strip()
 845
 846         playlist_description = json_ld_info.get(
 847             'description') or self._og_search_description(webpage, default=None)
 848
 849         if not timestamp:
 850             timestamp = parse_iso8601(self._search_regex(
 851                 [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
 852                  r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
 853                  r'"datePublished":\s*"([^"]+)'],
 854                 webpage, 'date', default=None))
 855
 856         entries = []
 857
 858         # article with multiple videos embedded with playlist.sxml (e.g.
 859         # http://www.bbc.com/sport/0/football/34475836)
 860         playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
 861         playlists.extend(re.findall(r'data-media-id="([^"]+/playlist\.sxml)"', webpage))
 862         if playlists:
 863             entries = [
 864                 self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
 865                 for playlist_url in playlists]
 866
 867         # news article with multiple videos embedded with data-playable
 868         data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
 869         if data_playables:
 870             for _, data_playable_json in data_playables:
 871                 data_playable = self._parse_json(
 872                     unescapeHTML(data_playable_json), playlist_id, fatal=False)
 873                 if not data_playable:
 874                     continue
 875                 settings = data_playable.get('settings', {})
 876                 if settings:
 877                     # data-playable with video vpid in settings.playlistObject.items (e.g.
 878                     # http://www.bbc.com/news/world-us-canada-34473351)
 879                     playlist_object = settings.get('playlistObject', {})
 880                     if playlist_object:
 881                         items = playlist_object.get('items')
 882                         if items and isinstance(items, list):
 883                             title = playlist_object['title']
 884                             description = playlist_object.get('summary')
 885                             duration = int_or_none(items[0].get('duration'))
 886                             programme_id = items[0].get('vpid')
 887                             formats, subtitles = self._download_media_selector(programme_id)
 888                             self._sort_formats(formats)
 889                             entries.append({
 890                                 'id': programme_id,
 891                                 'title': title,
 892                                 'description': description,
 893                                 'timestamp': timestamp,
 894                                 'duration': duration,
 895                                 'formats': formats,
 896                                 'subtitles': subtitles,
 897                             })
 898                     else:
 899                         # data-playable without vpid but with a playlist.sxml URLs
 900                         # in otherSettings.playlist (e.g.
 901                         # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
 902                         playlist = data_playable.get('otherSettings', {}).get('playlist', {})
 903                         if playlist:
 904                             entry = None
 905                             for key in ('streaming', 'progressiveDownload'):
 906                                 playlist_url = playlist.get('%sUrl' % key)
 907                                 if not playlist_url:
 908                                     continue
 909                                 try:
 910                                     info = self._extract_from_playlist_sxml(
 911                                         playlist_url, playlist_id, timestamp)
 912                                     if not entry:
 913                                         entry = info
 914                                     else:
 915                                         entry['title'] = info['title']
 916                                         entry['formats'].extend(info['formats'])
 917                                 except Exception as e:
 918                                     # Some playlist URL may fail with 500, at the same time
 919                                     # the other one may work fine (e.g.
 920                                     # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
 921                                     if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
 922                                         continue
 923                                     raise
 924                             if entry:
 925                                 self._sort_formats(entry['formats'])
 926                                 entries.append(entry)
 927
 928         if entries:
 929             return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 930
 931         # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
 932         programme_id = self._search_regex(
 933             [r'data-(?:video-player|media)-vpid="(%s)"' % self._ID_REGEX,
 934              r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
 935              r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
 936             webpage, 'vpid', default=None)
 937
 938         if programme_id:
 939             formats, subtitles = self._download_media_selector(programme_id)
 940             self._sort_formats(formats)
 941             # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
 942             digital_data = self._parse_json(
 943                 self._search_regex(
 944                     r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
 945                 programme_id, fatal=False)
 946             page_info = digital_data.get('page', {}).get('pageInfo', {})
 947             title = page_info.get('pageName') or self._og_search_title(webpage)
 948             description = page_info.get('description') or self._og_search_description(webpage)
 949             timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
 950             return {
 951                 'id': programme_id,
 952                 'title': title,
 953                 'description': description,
 954                 'timestamp': timestamp,
 955                 'formats': formats,
 956                 'subtitles': subtitles,
 957             }
 958
 959         # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
 960         # There are several setPayload calls may be present but the video
 961         # seems to be always related to the first one
 962         morph_payload = self._parse_json(
 963             self._search_regex(
 964                 r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
 965                 webpage, 'morph payload', default='{}'),
 966             playlist_id, fatal=False)
 967         if morph_payload:
 968             components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
 969             for component in components:
 970                 if not isinstance(component, dict):
 971                     continue
 972                 lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
 973                 if not lead_media:
 974                     continue
 975                 identifiers = lead_media.get('identifiers')
 976                 if not identifiers or not isinstance(identifiers, dict):
 977                     continue
 978                 programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
 979                 if not programme_id:
 980                     continue
 981                 title = lead_media.get('title') or self._og_search_title(webpage)
 982                 formats, subtitles = self._download_media_selector(programme_id)
 983                 self._sort_formats(formats)
 984                 description = lead_media.get('summary')
 985                 uploader = lead_media.get('masterBrand')
 986                 uploader_id = lead_media.get('mid')
 987                 duration = None
 988                 duration_d = lead_media.get('duration')
 989                 if isinstance(duration_d, dict):
 990                     duration = parse_duration(dict_get(
 991                         duration_d, ('rawDuration', 'formattedDuration', 'spokenDuration')))
 992                 return {
 993                     'id': programme_id,
 994                     'title': title,
 995                     'description': description,
 996                     'duration': duration,
 997                     'uploader': uploader,
 998                     'uploader_id': uploader_id,
 999                     'formats': formats,
1000                     'subtitles': subtitles,
1001                 }
1002
1003         bbc3_config = self._parse_json(
1004             self._search_regex(
1005                 r'(?s)bbcthreeConfig\s*=\s*({.+?})\s*;\s*<', webpage,
1006                 'bbcthree config', default='{}'),
1007             playlist_id, transform_source=js_to_json, fatal=False)
1008         if bbc3_config:
1009             bbc3_playlist = try_get(
1010                 bbc3_config, lambda x: x['payload']['content']['bbcMedia']['playlist'],
1011                 dict)
1012             if bbc3_playlist:
1013                 playlist_title = bbc3_playlist.get('title') or playlist_title
1014                 thumbnail = bbc3_playlist.get('holdingImageURL')
1015                 entries = []
1016                 for bbc3_item in bbc3_playlist['items']:
1017                     programme_id = bbc3_item.get('versionID')
1018                     if not programme_id:
1019                         continue
1020                     formats, subtitles = self._download_media_selector(programme_id)
1021                     self._sort_formats(formats)
1022                     entries.append({
1023                         'id': programme_id,
1024                         'title': playlist_title,
1025                         'thumbnail': thumbnail,
1026                         'timestamp': timestamp,
1027                         'formats': formats,
1028                         'subtitles': subtitles,
1029                     })
1030                 return self.playlist_result(
1031                     entries, playlist_id, playlist_title, playlist_description)
1032
1033         def extract_all(pattern):
1034             return list(filter(None, map(
1035                 lambda s: self._parse_json(s, playlist_id, fatal=False),
1036                 re.findall(pattern, webpage))))
1037
1038         # Multiple video article (e.g.
1039         # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
1040         EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX
1041         entries = []
1042         for match in extract_all(r'new\s+SMP\(({.+?})\)'):
1043             embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
1044             if embed_url and re.match(EMBED_URL, embed_url):
1045                 entries.append(embed_url)
1046         entries.extend(re.findall(
1047             r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
1048         if entries:
1049             return self.playlist_result(
1050                 [self.url_result(entry_, 'BBCCoUk') for entry_ in entries],
1051                 playlist_id, playlist_title, playlist_description)
1052
1053         # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
1054         medias = extract_all(r"data-media-meta='({[^']+})'")
1055
1056         if not medias:
1057             # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
1058             media_asset = self._search_regex(
1059                 r'mediaAssetPage\.init\(\s*({.+?}), "/',
1060                 webpage, 'media asset', default=None)
1061             if media_asset:
1062                 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
1063                 medias = []
1064                 for video in media_asset_page.get('videos', {}).values():
1065                     medias.extend(video.values())
1066
1067         if not medias:
1068             # Multiple video playlist with single `now playing` entry (e.g.
1069             # http://www.bbc.com/news/video_and_audio/must_see/33767813)
1070             vxp_playlist = self._parse_json(
1071                 self._search_regex(
1072                     r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>',
1073                     webpage, 'playlist data'),
1074                 playlist_id)
1075             playlist_medias = []
1076             for item in vxp_playlist:
1077                 media = item.get('media')
1078                 if not media:
1079                     continue
1080                 playlist_medias.append(media)
1081                 # Download single video if found media with asset id matching the video id from URL
1082                 if item.get('advert', {}).get('assetId') == playlist_id:
1083                     medias = [media]
1084                     break
1085             # Fallback to the whole playlist
1086             if not medias:
1087                 medias = playlist_medias
1088
1089         entries = []
1090         for num, media_meta in enumerate(medias, start=1):
1091             formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
1092             if not formats:
1093                 continue
1094             self._sort_formats(formats)
1095
1096             video_id = media_meta.get('externalId')
1097             if not video_id:
1098                 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
1099
1100             title = media_meta.get('caption')
1101             if not title:
1102                 title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
1103
1104             duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
1105
1106             images = []
1107             for image in media_meta.get('images', {}).values():
1108                 images.extend(image.values())
1109             if 'image' in media_meta:
1110                 images.append(media_meta['image'])
1111
1112             thumbnails = [{
1113                 'url': image.get('href'),
1114                 'width': int_or_none(image.get('width')),
1115                 'height': int_or_none(image.get('height')),
1116             } for image in images]
1117
1118             entries.append({
1119                 'id': video_id,
1120                 'title': title,
1121                 'thumbnails': thumbnails,
1122                 'duration': duration,
1123                 'timestamp': timestamp,
1124                 'formats': formats,
1125                 'subtitles': subtitles,
1126             })
1127
1128         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
1129
1130
1131 class BBCCoUkArticleIE(InfoExtractor):
1132     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
1133     IE_NAME = 'bbc.co.uk:article'
1134     IE_DESC = 'BBC articles'
1135
1136     _TEST = {
1137         'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer',
1138         'info_dict': {
1139             'id': '3jNQLTMrPlYGTBn0WV6M2MS',
1140             'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four',
1141             'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.',
1142         },
1143         'playlist_count': 4,
1144         'add_ie': ['BBCCoUk'],
1145     }
1146
1147     def _real_extract(self, url):
1148         playlist_id = self._match_id(url)
1149
1150         webpage = self._download_webpage(url, playlist_id)
1151
1152         title = self._og_search_title(webpage)
1153         description = self._og_search_description(webpage).strip()
1154
1155         entries = [self.url_result(programme_url) for programme_url in re.findall(
1156             r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
1157
1158         return self.playlist_result(entries, playlist_id, title, description)
1159
1160
1161 class BBCCoUkPlaylistBaseIE(InfoExtractor):
1162     def _entries(self, webpage, url, playlist_id):
1163         single_page = 'page' in compat_urlparse.parse_qs(
1164             compat_urlparse.urlparse(url).query)
1165         for page_num in itertools.count(2):
1166             for video_id in re.findall(
1167                     self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage):
1168                 yield self.url_result(
1169                     self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
1170             if single_page:
1171                 return
1172             next_page = self._search_regex(
1173                 r'<li[^>]+class=(["\'])pagination_+next\1[^>]*><a[^>]+href=(["\'])(?P<url>(?:(?!\2).)+)\2',
1174                 webpage, 'next page url', default=None, group='url')
1175             if not next_page:
1176                 break
1177             webpage = self._download_webpage(
1178                 compat_urlparse.urljoin(url, next_page), playlist_id,
1179                 'Downloading page %d' % page_num, page_num)
1180
1181     def _real_extract(self, url):
1182         playlist_id = self._match_id(url)
1183
1184         webpage = self._download_webpage(url, playlist_id)
1185
1186         title, description = self._extract_title_and_description(webpage)
1187
1188         return self.playlist_result(
1189             self._entries(webpage, url, playlist_id),
1190             playlist_id, title, description)
1191
1192
1193 class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
1194     IE_NAME = 'bbc.co.uk:iplayer:playlist'
1195     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
1196     _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
1197     _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
1198     _TESTS = [{
1199         'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v',
1200         'info_dict': {
1201             'id': 'b05rcz9v',
1202             'title': 'The Disappearance',
1203             'description': 'French thriller serial about a missing teenager.',
1204         },
1205         'playlist_mincount': 6,
1206         'skip': 'This programme is not currently available on BBC iPlayer',
1207     }, {
1208         # Available for over a year unlike 30 days for most other programmes
1209         'url': 'http://www.bbc.co.uk/iplayer/group/p02tcc32',
1210         'info_dict': {
1211             'id': 'p02tcc32',
1212             'title': 'Bohemian Icons',
1213             'description': 'md5:683e901041b2fe9ba596f2ab04c4dbe7',
1214         },
1215         'playlist_mincount': 10,
1216     }]
1217
1218     def _extract_title_and_description(self, webpage):
1219         title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
1220         description = self._search_regex(
1221             r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
1222             webpage, 'description', fatal=False, group='value')
1223         return title, description
1224
1225
1226 class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
1227     IE_NAME = 'bbc.co.uk:playlist'
1228     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
1229     _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
1230     _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
1231     _TESTS = [{
1232         'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1233         'info_dict': {
1234             'id': 'b05rcz9v',
1235             'title': 'The Disappearance - Clips - BBC Four',
1236             'description': 'French thriller serial about a missing teenager.',
1237         },
1238         'playlist_mincount': 7,
1239     }, {
1240         # multipage playlist, explicit page
1241         'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips?page=1',
1242         'info_dict': {
1243             'id': 'b00mfl7n',
1244             'title': 'Frozen Planet - Clips - BBC One',
1245             'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1246         },
1247         'playlist_mincount': 24,
1248     }, {
1249         # multipage playlist, all pages
1250         'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips',
1251         'info_dict': {
1252             'id': 'b00mfl7n',
1253             'title': 'Frozen Planet - Clips - BBC One',
1254             'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1255         },
1256         'playlist_mincount': 142,
1257     }, {
1258         'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/broadcasts/2016/06',
1259         'only_matching': True,
1260     }, {
1261         'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1262         'only_matching': True,
1263     }, {
1264         'url': 'http://www.bbc.co.uk/programmes/b055jkys/episodes/player',
1265         'only_matching': True,
1266     }]
1267
1268     def _extract_title_and_description(self, webpage):
1269         title = self._og_search_title(webpage, fatal=False)
1270         description = self._og_search_description(webpage)
1271         return title, description