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