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