2 from __future__ import unicode_literals
7 from .common import InfoExtractor
22 from ..compat import (
23 compat_etree_fromstring,
29 class BBCCoUkIE(InfoExtractor):
31 IE_DESC = 'BBC iPlayer'
32 _ID_REGEX = r'[pbw][\da-z]{7}'
35 (?:www\.)?bbc\.co\.uk/
37 programmes/(?!articles/)|
38 iplayer(?:/[^/]+)?/(?:episode/|playlist/)|
39 music/(?:clips|audiovideo/popular)[/#]|
41 events/[^/]+/play/[^/]+/
43 (?P<id>%s)(?!/(?:episodes|broadcasts|clips))
46 _LOGIN_URL = 'https://account.bbc.com/signin'
47 _NETRC_MACHINE = 'bbc'
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',
57 _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
58 _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
67 'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
71 'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
72 'description': 'The Canadian poet and songwriter reflects on his musical career.',
76 'skip_download': True,
80 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
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.",
90 'skip_download': True,
92 'skip': 'Episode is no longer available on BBC iPlayer Radio',
95 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
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.',
105 'skip_download': True,
107 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
110 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
114 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
115 'description': '2. Invasion',
120 'skip_download': True,
122 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
124 'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
128 'title': 'Pete Tong, The Essential New Tune Special',
129 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
134 'skip_download': True,
136 'skip': 'Episode is no longer available on BBC iPlayer Radio',
138 'url': 'http://www.bbc.co.uk/music/clips/p022h44b',
143 'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
144 'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
149 'skip_download': True,
152 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
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',
163 'skip_download': True,
166 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
170 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
171 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
176 'skip_download': True,
178 'skip': 'geolocation',
180 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
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',
190 'skip_download': True,
192 'skip': 'geolocation',
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',
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!",
205 'skip_download': True,
207 'skip': 'Now it\'s really geo-restricted',
209 # compact player (https://github.com/rg3/youtube-dl/issues/8147)
210 'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
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',
219 'skip_download': True,
222 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
223 'only_matching': True,
225 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
226 'only_matching': True,
228 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
229 'only_matching': True,
231 'url': 'http://www.bbc.co.uk/radio/player/p03cchwf',
232 'only_matching': True,
234 'url': 'https://www.bbc.co.uk/music/audiovideo/popular#p055bc55',
235 'only_matching': True,
237 'url': 'http://www.bbc.co.uk/programmes/w3csv1y9',
238 'only_matching': True,
241 _USP_RE = r'/([^/]+?)\.ism(?:\.hlsv2\.ism)?/[^/]+\.m3u8'
244 username, password = self._get_login_info()
248 login_page = self._download_webpage(
249 self._LOGIN_URL, None, 'Downloading signin page')
251 login_form = self._hidden_inputs(login_page)
254 'username': username,
255 'password': password,
258 post_url = urljoin(self._LOGIN_URL, self._search_regex(
259 r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
260 'post url', default=self._LOGIN_URL, group='url'))
262 response, urlh = self._download_webpage_handle(
263 post_url, None, 'Logging in', data=urlencode_postdata(login_form),
264 headers={'Referer': self._LOGIN_URL})
266 if self._LOGIN_URL in urlh.geturl():
267 error = clean_html(get_element_by_class('form-message', response))
269 raise ExtractorError(
270 'Unable to login: %s' % error, expected=True)
271 raise ExtractorError('Unable to log in')
273 def _real_initialize(self):
276 class MediaSelectionError(Exception):
277 def __init__(self, id):
280 def _extract_asx_playlist(self, connection, programme_id):
281 asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
282 return [ref.get('href') for ref in asx.findall('./Entry/ref')]
284 def _extract_items(self, playlist):
285 return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
287 def _findall_ns(self, element, xpath):
289 for ns in self._NAMESPACES:
290 elements.extend(element.findall(xpath % ns))
293 def _extract_medias(self, media_selection):
294 error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS)
296 media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS)
297 if error is not None:
298 raise BBCCoUkIE.MediaSelectionError(error.get('id'))
299 return self._findall_ns(media_selection, './{%s}media')
301 def _extract_connections(self, media):
302 return self._findall_ns(media, './{%s}connection')
304 def _get_subtitles(self, media, programme_id):
306 for connection in self._extract_connections(media):
307 captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
308 lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
311 'url': connection.get('href'),
317 def _raise_extractor_error(self, media_selection_error):
318 raise ExtractorError(
319 '%s returned error: %s' % (self.IE_NAME, media_selection_error.id),
322 def _download_media_selector(self, programme_id):
323 last_exception = None
324 for mediaselector_url in self._MEDIASELECTOR_URLS:
326 return self._download_media_selector_url(
327 mediaselector_url % programme_id, programme_id)
328 except BBCCoUkIE.MediaSelectionError as e:
329 if e.id in ('notukerror', 'geolocation', 'selectionunavailable'):
332 self._raise_extractor_error(e)
333 self._raise_extractor_error(last_exception)
335 def _download_media_selector_url(self, url, programme_id=None):
337 media_selection = self._download_xml(
338 url, programme_id, 'Downloading media selection XML')
339 except ExtractorError as ee:
340 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code in (403, 404):
341 media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8'))
344 return self._process_media_selector(media_selection, programme_id)
346 def _process_media_selector(self, media_selection, programme_id):
351 for media in self._extract_medias(media_selection):
352 kind = media.get('kind')
353 if kind in ('video', 'audio'):
354 bitrate = int_or_none(media.get('bitrate'))
355 encoding = media.get('encoding')
356 service = media.get('service')
357 width = int_or_none(media.get('width'))
358 height = int_or_none(media.get('height'))
359 file_size = int_or_none(media.get('media_file_size'))
360 for connection in self._extract_connections(media):
361 href = connection.get('href')
366 conn_kind = connection.get('kind')
367 protocol = connection.get('protocol')
368 supplier = connection.get('supplier')
369 transfer_format = connection.get('transferFormat')
370 format_id = supplier or conn_kind or protocol
372 format_id = '%s_%s' % (service, format_id)
374 if supplier == 'asx':
375 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
378 'format_id': 'ref%s_%s' % (i, format_id),
380 elif transfer_format == 'dash':
381 formats.extend(self._extract_mpd_formats(
382 href, programme_id, mpd_id=format_id, fatal=False))
383 elif transfer_format == 'hls':
384 formats.extend(self._extract_m3u8_formats(
385 href, programme_id, ext='mp4', entry_protocol='m3u8_native',
386 m3u8_id=format_id, fatal=False))
387 if re.search(self._USP_RE, href):
388 usp_formats = self._extract_m3u8_formats(
389 re.sub(self._USP_RE, r'/\1.ism/\1.m3u8', href),
390 programme_id, ext='mp4', entry_protocol='m3u8_native',
391 m3u8_id=format_id, fatal=False)
392 for f in usp_formats:
393 if f.get('height') and f['height'] > 720:
396 elif transfer_format == 'hds':
397 formats.extend(self._extract_f4m_formats(
398 href, programme_id, f4m_id=format_id, fatal=False))
400 if not service and not supplier and bitrate:
401 format_id += '-%d' % bitrate
403 'format_id': format_id,
404 'filesize': file_size,
419 if protocol in ('http', 'https'):
424 elif protocol == 'rtmp':
425 application = connection.get('application', 'ondemand')
426 auth_string = connection.get('authString')
427 identifier = connection.get('identifier')
428 server = connection.get('server')
430 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
431 'play_path': identifier,
432 'app': '%s?%s' % (application, auth_string),
433 'page_url': 'http://www.bbc.co.uk',
434 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
441 elif kind == 'captions':
442 subtitles = self.extract_subtitles(media, programme_id)
443 return formats, subtitles
445 def _download_playlist(self, playlist_id):
447 playlist = self._download_json(
448 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
449 playlist_id, 'Downloading playlist JSON')
451 version = playlist.get('defaultAvailableVersion')
453 smp_config = version['smpConfig']
454 title = smp_config['title']
455 description = smp_config['summary']
456 for item in smp_config['items']:
458 if kind not in ('programme', 'radioProgramme'):
460 programme_id = item.get('vpid')
461 duration = int_or_none(item.get('duration'))
462 formats, subtitles = self._download_media_selector(programme_id)
463 return programme_id, title, description, duration, formats, subtitles
464 except ExtractorError as ee:
465 if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
468 # fallback to legacy playlist
469 return self._process_legacy_playlist(playlist_id)
471 def _process_legacy_playlist_url(self, url, display_id):
472 playlist = self._download_legacy_playlist_url(url, display_id)
473 return self._extract_from_legacy_playlist(playlist, display_id)
475 def _process_legacy_playlist(self, playlist_id):
476 return self._process_legacy_playlist_url(
477 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
479 def _download_legacy_playlist_url(self, url, playlist_id=None):
480 return self._download_xml(
481 url, playlist_id, 'Downloading legacy playlist XML')
483 def _extract_from_legacy_playlist(self, playlist, playlist_id):
484 no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS)
485 if no_items is not None:
486 reason = no_items.get('reason')
487 if reason == 'preAvailability':
488 msg = 'Episode %s is not yet available' % playlist_id
489 elif reason == 'postAvailability':
490 msg = 'Episode %s is no longer available' % playlist_id
491 elif reason == 'noMedia':
492 msg = 'Episode %s is not currently available' % playlist_id
494 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
495 raise ExtractorError(msg, expected=True)
497 for item in self._extract_items(playlist):
498 kind = item.get('kind')
499 if kind not in ('programme', 'radioProgramme'):
501 title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
502 description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
503 description = description_el.text if description_el is not None else None
505 def get_programme_id(item):
506 def get_from_attributes(item):
507 for p in('identifier', 'group'):
509 if value and re.match(r'^[pb][\da-z]{7}$', value):
511 get_from_attributes(item)
512 mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS)
513 if mediator is not None:
514 return get_from_attributes(mediator)
516 programme_id = get_programme_id(item)
517 duration = int_or_none(item.get('duration'))
520 formats, subtitles = self._download_media_selector(programme_id)
522 formats, subtitles = self._process_media_selector(item, playlist_id)
523 programme_id = playlist_id
525 return programme_id, title, description, duration, formats, subtitles
527 def _real_extract(self, url):
528 group_id = self._match_id(url)
530 webpage = self._download_webpage(url, group_id, 'Downloading video page')
532 error = self._search_regex(
533 r'<div\b[^>]+\bclass=["\']smp__message delta["\'][^>]*>([^<]+)<',
534 webpage, 'error', default=None)
536 raise ExtractorError(error, expected=True)
541 tviplayer = self._search_regex(
542 r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
543 webpage, 'player', default=None)
546 player = self._parse_json(tviplayer, group_id).get('player', {})
547 duration = int_or_none(player.get('duration'))
548 programme_id = player.get('vpid')
551 programme_id = self._search_regex(
552 r'"vpid"\s*:\s*"(%s)"' % self._ID_REGEX, webpage, 'vpid', fatal=False, default=None)
555 formats, subtitles = self._download_media_selector(programme_id)
556 title = self._og_search_title(webpage, default=None) or self._html_search_regex(
557 (r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>',
558 r'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage, 'title')
559 description = self._search_regex(
560 (r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
561 r'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'),
562 webpage, 'description', default=None)
564 description = self._html_search_meta('description', webpage)
566 programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
568 self._sort_formats(formats)
573 'description': description,
574 'thumbnail': self._og_search_thumbnail(webpage, default=None),
575 'duration': duration,
577 'subtitles': subtitles,
581 class BBCIE(BBCCoUkIE):
584 _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
586 _MEDIASELECTOR_URLS = [
587 # Provides HQ HLS streams but fails with geolocation in some cases when it's
588 # even not geo restricted at all
589 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
590 # Provides more formats, namely direct mp4 links, but fails on some videos with
591 # notukerror for non UK (?) users (e.g.
592 # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
593 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
594 # Provides fewer formats, but works everywhere for everybody (hopefully)
595 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
599 # article with multiple videos embedded with data-playable containing vpids
600 'url': 'http://www.bbc.com/news/world-europe-32668511',
602 'id': 'world-europe-32668511',
603 'title': 'Russia stages massive WW2 parade despite Western boycott',
604 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
608 # article with multiple videos embedded with data-playable (more videos)
609 'url': 'http://www.bbc.com/news/business-28299555',
611 'id': 'business-28299555',
612 'title': 'Farnborough Airshow: Video highlights',
613 'description': 'BBC reports and video highlights at the Farnborough Airshow.',
618 # article with multiple videos embedded with `new SMP()`
620 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
622 'id': '3662a707-0af9-3149-963f-47bea720b460',
625 'playlist_count': 18,
627 # single video embedded with data-playable containing vpid
628 'url': 'http://www.bbc.com/news/world-europe-32041533',
632 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
633 'description': 'md5:2868290467291b37feda7863f7a83f54',
635 'timestamp': 1427219242,
636 'upload_date': '20150324',
640 'skip_download': True,
643 # article with single video embedded with data-playable containing XML playlist
644 # with direct video links as progressiveDownloadUrl (for now these are extracted)
645 # and playlist with f4m and m3u8 as streamingUrl
646 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
648 'id': '150615_telabyad_kentin_cogu',
650 'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
651 'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
652 'timestamp': 1434397334,
653 'upload_date': '20150615',
656 'skip_download': True,
659 # single video embedded with data-playable containing XML playlists (regional section)
660 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
662 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
664 'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
665 'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
666 'timestamp': 1434713142,
667 'upload_date': '20150619',
670 'skip_download': True,
673 # single video from video playlist embedded with vxp-playlist-data JSON
674 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
678 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
680 'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
683 'skip_download': True,
686 # single video story with digitalData
687 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
691 'title': 'Sri Lanka’s spicy secret',
692 'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
693 'timestamp': 1437674293,
694 'upload_date': '20150723',
698 'skip_download': True,
701 # single video story without digitalData
702 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
706 'title': 'Hyundai Santa Fe Sport: Rock star',
707 'description': 'md5:b042a26142c4154a6e472933cf20793d',
708 'timestamp': 1415867444,
709 'upload_date': '20141113',
713 'skip_download': True,
716 # single video embedded with Morph
717 'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
721 'title': "Nigeria v Japan - Men's First Round",
722 'description': 'Live coverage of the first round from Group B at the Amazonia Arena.',
724 'uploader': 'BBC Sport',
725 'uploader_id': 'bbc_sport',
729 'skip_download': True,
731 'skip': 'Georestricted to UK',
733 # single video with playlist.sxml URL in playlist param
734 'url': 'http://www.bbc.com/sport/0/football/33653409',
738 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
739 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
744 'skip_download': True,
747 # article with multiple videos embedded with playlist.sxml in playlist param
748 'url': 'http://www.bbc.com/sport/0/football/34475836',
751 'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
752 'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
756 # school report article with single video
757 'url': 'http://www.bbc.co.uk/schoolreport/35744779',
760 'title': 'School which breaks down barriers in Jerusalem',
764 # single video with playlist URL from weather section
765 'url': 'http://www.bbc.com/weather/features/33601775',
766 'only_matching': True,
768 # custom redirection to www.bbc.com
769 'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
770 'only_matching': True,
772 # single video article embedded with data-media-vpid
773 'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
774 'only_matching': True,
778 def suitable(cls, url):
779 EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE)
780 return (False if any(ie.suitable(url) for ie in EXCLUDE_IE)
781 else super(BBCIE, cls).suitable(url))
783 def _extract_from_media_meta(self, media_meta, video_id):
784 # Direct links to media in media metadata (e.g.
785 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
786 # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
787 source_files = media_meta.get('sourceFiles')
791 'format_id': format_id,
792 'ext': f.get('encoding'),
793 'tbr': float_or_none(f.get('bitrate'), 1000),
794 'filesize': int_or_none(f.get('filesize')),
795 } for format_id, f in source_files.items() if f.get('url')], []
797 programme_id = media_meta.get('externalId')
799 return self._download_media_selector(programme_id)
801 # Process playlist.sxml as legacy playlist
802 href = media_meta.get('href')
804 playlist = self._download_legacy_playlist_url(href)
805 _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
806 return formats, subtitles
810 def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
811 programme_id, title, description, duration, formats, subtitles = \
812 self._process_legacy_playlist_url(url, playlist_id)
813 self._sort_formats(formats)
817 'description': description,
818 'duration': duration,
819 'timestamp': timestamp,
821 'subtitles': subtitles,
824 def _real_extract(self, url):
825 playlist_id = self._match_id(url)
827 webpage = self._download_webpage(url, playlist_id)
829 json_ld_info = self._search_json_ld(webpage, playlist_id, default={})
830 timestamp = json_ld_info.get('timestamp')
832 playlist_title = json_ld_info.get('title')
833 if not playlist_title:
834 playlist_title = self._og_search_title(
835 webpage, default=None) or self._html_search_regex(
836 r'<title>(.+?)</title>', webpage, 'playlist title', default=None)
838 playlist_title = re.sub(r'(.+)\s*-\s*BBC.*?$', r'\1', playlist_title).strip()
840 playlist_description = json_ld_info.get(
841 'description') or self._og_search_description(webpage, default=None)
844 timestamp = parse_iso8601(self._search_regex(
845 [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
846 r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
847 r'"datePublished":\s*"([^"]+)'],
848 webpage, 'date', default=None))
852 # article with multiple videos embedded with playlist.sxml (e.g.
853 # http://www.bbc.com/sport/0/football/34475836)
854 playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
855 playlists.extend(re.findall(r'data-media-id="([^"]+/playlist\.sxml)"', webpage))
858 self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
859 for playlist_url in playlists]
861 # news article with multiple videos embedded with data-playable
862 data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
864 for _, data_playable_json in data_playables:
865 data_playable = self._parse_json(
866 unescapeHTML(data_playable_json), playlist_id, fatal=False)
867 if not data_playable:
869 settings = data_playable.get('settings', {})
871 # data-playable with video vpid in settings.playlistObject.items (e.g.
872 # http://www.bbc.com/news/world-us-canada-34473351)
873 playlist_object = settings.get('playlistObject', {})
875 items = playlist_object.get('items')
876 if items and isinstance(items, list):
877 title = playlist_object['title']
878 description = playlist_object.get('summary')
879 duration = int_or_none(items[0].get('duration'))
880 programme_id = items[0].get('vpid')
881 formats, subtitles = self._download_media_selector(programme_id)
882 self._sort_formats(formats)
886 'description': description,
887 'timestamp': timestamp,
888 'duration': duration,
890 'subtitles': subtitles,
893 # data-playable without vpid but with a playlist.sxml URLs
894 # in otherSettings.playlist (e.g.
895 # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
896 playlist = data_playable.get('otherSettings', {}).get('playlist', {})
899 for key in ('streaming', 'progressiveDownload'):
900 playlist_url = playlist.get('%sUrl' % key)
904 info = self._extract_from_playlist_sxml(
905 playlist_url, playlist_id, timestamp)
909 entry['title'] = info['title']
910 entry['formats'].extend(info['formats'])
911 except Exception as e:
912 # Some playlist URL may fail with 500, at the same time
913 # the other one may work fine (e.g.
914 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
915 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
919 self._sort_formats(entry['formats'])
920 entries.append(entry)
923 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
925 # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
926 programme_id = self._search_regex(
927 [r'data-(?:video-player|media)-vpid="(%s)"' % self._ID_REGEX,
928 r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
929 r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
930 webpage, 'vpid', default=None)
933 formats, subtitles = self._download_media_selector(programme_id)
934 self._sort_formats(formats)
935 # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
936 digital_data = self._parse_json(
938 r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
939 programme_id, fatal=False)
940 page_info = digital_data.get('page', {}).get('pageInfo', {})
941 title = page_info.get('pageName') or self._og_search_title(webpage)
942 description = page_info.get('description') or self._og_search_description(webpage)
943 timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
947 'description': description,
948 'timestamp': timestamp,
950 'subtitles': subtitles,
953 # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
954 # There are several setPayload calls may be present but the video
955 # seems to be always related to the first one
956 morph_payload = self._parse_json(
958 r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
959 webpage, 'morph payload', default='{}'),
960 playlist_id, fatal=False)
962 components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
963 for component in components:
964 if not isinstance(component, dict):
966 lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
969 identifiers = lead_media.get('identifiers')
970 if not identifiers or not isinstance(identifiers, dict):
972 programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
975 title = lead_media.get('title') or self._og_search_title(webpage)
976 formats, subtitles = self._download_media_selector(programme_id)
977 self._sort_formats(formats)
978 description = lead_media.get('summary')
979 uploader = lead_media.get('masterBrand')
980 uploader_id = lead_media.get('mid')
982 duration_d = lead_media.get('duration')
983 if isinstance(duration_d, dict):
984 duration = parse_duration(dict_get(
985 duration_d, ('rawDuration', 'formattedDuration', 'spokenDuration')))
989 'description': description,
990 'duration': duration,
991 'uploader': uploader,
992 'uploader_id': uploader_id,
994 'subtitles': subtitles,
997 def extract_all(pattern):
998 return list(filter(None, map(
999 lambda s: self._parse_json(s, playlist_id, fatal=False),
1000 re.findall(pattern, webpage))))
1002 # Multiple video article (e.g.
1003 # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
1004 EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX
1006 for match in extract_all(r'new\s+SMP\(({.+?})\)'):
1007 embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
1008 if embed_url and re.match(EMBED_URL, embed_url):
1009 entries.append(embed_url)
1010 entries.extend(re.findall(
1011 r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
1013 return self.playlist_result(
1014 [self.url_result(entry_, 'BBCCoUk') for entry_ in entries],
1015 playlist_id, playlist_title, playlist_description)
1017 # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
1018 medias = extract_all(r"data-media-meta='({[^']+})'")
1021 # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
1022 media_asset = self._search_regex(
1023 r'mediaAssetPage\.init\(\s*({.+?}), "/',
1024 webpage, 'media asset', default=None)
1026 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
1028 for video in media_asset_page.get('videos', {}).values():
1029 medias.extend(video.values())
1032 # Multiple video playlist with single `now playing` entry (e.g.
1033 # http://www.bbc.com/news/video_and_audio/must_see/33767813)
1034 vxp_playlist = self._parse_json(
1036 r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>',
1037 webpage, 'playlist data'),
1039 playlist_medias = []
1040 for item in vxp_playlist:
1041 media = item.get('media')
1044 playlist_medias.append(media)
1045 # Download single video if found media with asset id matching the video id from URL
1046 if item.get('advert', {}).get('assetId') == playlist_id:
1049 # Fallback to the whole playlist
1051 medias = playlist_medias
1054 for num, media_meta in enumerate(medias, start=1):
1055 formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
1058 self._sort_formats(formats)
1060 video_id = media_meta.get('externalId')
1062 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
1064 title = media_meta.get('caption')
1066 title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
1068 duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
1071 for image in media_meta.get('images', {}).values():
1072 images.extend(image.values())
1073 if 'image' in media_meta:
1074 images.append(media_meta['image'])
1077 'url': image.get('href'),
1078 'width': int_or_none(image.get('width')),
1079 'height': int_or_none(image.get('height')),
1080 } for image in images]
1085 'thumbnails': thumbnails,
1086 'duration': duration,
1087 'timestamp': timestamp,
1089 'subtitles': subtitles,
1092 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
1095 class BBCCoUkArticleIE(InfoExtractor):
1096 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
1097 IE_NAME = 'bbc.co.uk:article'
1098 IE_DESC = 'BBC articles'
1101 'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer',
1103 'id': '3jNQLTMrPlYGTBn0WV6M2MS',
1104 'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four',
1105 'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.',
1107 'playlist_count': 4,
1108 'add_ie': ['BBCCoUk'],
1111 def _real_extract(self, url):
1112 playlist_id = self._match_id(url)
1114 webpage = self._download_webpage(url, playlist_id)
1116 title = self._og_search_title(webpage)
1117 description = self._og_search_description(webpage).strip()
1119 entries = [self.url_result(programme_url) for programme_url in re.findall(
1120 r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
1122 return self.playlist_result(entries, playlist_id, title, description)
1125 class BBCCoUkPlaylistBaseIE(InfoExtractor):
1126 def _entries(self, webpage, url, playlist_id):
1127 single_page = 'page' in compat_urlparse.parse_qs(
1128 compat_urlparse.urlparse(url).query)
1129 for page_num in itertools.count(2):
1130 for video_id in re.findall(
1131 self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage):
1132 yield self.url_result(
1133 self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
1136 next_page = self._search_regex(
1137 r'<li[^>]+class=(["\'])pagination_+next\1[^>]*><a[^>]+href=(["\'])(?P<url>(?:(?!\2).)+)\2',
1138 webpage, 'next page url', default=None, group='url')
1141 webpage = self._download_webpage(
1142 compat_urlparse.urljoin(url, next_page), playlist_id,
1143 'Downloading page %d' % page_num, page_num)
1145 def _real_extract(self, url):
1146 playlist_id = self._match_id(url)
1148 webpage = self._download_webpage(url, playlist_id)
1150 title, description = self._extract_title_and_description(webpage)
1152 return self.playlist_result(
1153 self._entries(webpage, url, playlist_id),
1154 playlist_id, title, description)
1157 class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
1158 IE_NAME = 'bbc.co.uk:iplayer:playlist'
1159 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
1160 _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
1161 _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
1163 'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v',
1166 'title': 'The Disappearance',
1167 'description': 'French thriller serial about a missing teenager.',
1169 'playlist_mincount': 6,
1170 'skip': 'This programme is not currently available on BBC iPlayer',
1172 # Available for over a year unlike 30 days for most other programmes
1173 'url': 'http://www.bbc.co.uk/iplayer/group/p02tcc32',
1176 'title': 'Bohemian Icons',
1177 'description': 'md5:683e901041b2fe9ba596f2ab04c4dbe7',
1179 'playlist_mincount': 10,
1182 def _extract_title_and_description(self, webpage):
1183 title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
1184 description = self._search_regex(
1185 r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
1186 webpage, 'description', fatal=False, group='value')
1187 return title, description
1190 class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
1191 IE_NAME = 'bbc.co.uk:playlist'
1192 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
1193 _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
1194 _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
1196 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1199 'title': 'The Disappearance - Clips - BBC Four',
1200 'description': 'French thriller serial about a missing teenager.',
1202 'playlist_mincount': 7,
1204 # multipage playlist, explicit page
1205 'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips?page=1',
1208 'title': 'Frozen Planet - Clips - BBC One',
1209 'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1211 'playlist_mincount': 24,
1213 # multipage playlist, all pages
1214 'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips',
1217 'title': 'Frozen Planet - Clips - BBC One',
1218 'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1220 'playlist_mincount': 142,
1222 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/broadcasts/2016/06',
1223 'only_matching': True,
1225 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1226 'only_matching': True,
1228 'url': 'http://www.bbc.co.uk/programmes/b055jkys/episodes/player',
1229 'only_matching': True,
1232 def _extract_title_and_description(self, webpage):
1233 title = self._og_search_title(webpage, fatal=False)
1234 description = self._og_search_description(webpage)
1235 return title, description