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