2 from __future__ import unicode_literals
7 from .common import InfoExtractor
23 from ..compat import (
24 compat_etree_fromstring,
30 class BBCCoUkIE(InfoExtractor):
32 IE_DESC = 'BBC iPlayer'
33 _ID_REGEX = r'[pbw][\da-z]{7}'
36 (?:www\.)?bbc\.co\.uk/
38 programmes/(?!articles/)|
39 iplayer(?:/[^/]+)?/(?:episode/|playlist/)|
40 music/(?:clips|audiovideo/popular)[/#]|
42 events/[^/]+/play/[^/]+/
44 (?P<id>%s)(?!/(?:episodes|broadcasts|clips))
47 _LOGIN_URL = 'https://account.bbc.com/signin'
48 _NETRC_MACHINE = 'bbc'
50 _MEDIASELECTOR_URLS = [
51 # Provides HQ HLS streams with even better quality that pc mediaset but fails
52 # with geolocation in some cases when it's even not geo restricted at all (e.g.
53 # http://www.bbc.co.uk/programmes/b06bp7lf). Also may fail with selectionunavailable.
54 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
55 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
58 _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
59 _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
68 'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
72 'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
73 'description': 'The Canadian poet and songwriter reflects on his musical career.',
77 'skip_download': True,
81 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
85 'title': 'The Man in Black: Series 3: The Printed Name',
86 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
91 'skip_download': True,
93 'skip': 'Episode is no longer available on BBC iPlayer Radio',
96 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
100 'title': 'The Voice UK: Series 3: Blind Auditions 5',
101 '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.',
106 'skip_download': True,
108 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
111 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
115 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
116 'description': '2. Invasion',
121 'skip_download': True,
123 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
125 'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
129 'title': 'Pete Tong, The Essential New Tune Special',
130 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
135 'skip_download': True,
137 'skip': 'Episode is no longer available on BBC iPlayer Radio',
139 'url': 'http://www.bbc.co.uk/music/clips/p022h44b',
144 'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
145 'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
150 'skip_download': True,
153 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
158 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
159 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
164 'skip_download': True,
167 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
171 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
172 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
177 'skip_download': True,
179 'skip': 'geolocation',
181 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
185 '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.',
186 'title': 'Royal Academy Summer Exhibition',
191 'skip_download': True,
193 'skip': 'geolocation',
195 # iptv-all mediaset fails with geolocation however there is no geo restriction
196 # for this programme at all
197 'url': 'http://www.bbc.co.uk/programmes/b06rkn85',
201 'title': "Best of the Mini-Mixes 2015: Part 3, Annie Mac's Friday Night - BBC Radio 1",
202 'description': "Annie has part three in the Best of the Mini-Mixes 2015, plus the year's Most Played!",
206 'skip_download': True,
208 'skip': 'Now it\'s really geo-restricted',
210 # compact player (https://github.com/rg3/youtube-dl/issues/8147)
211 'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
215 'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
216 'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
220 'skip_download': True,
223 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
224 'only_matching': True,
226 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
227 'only_matching': True,
229 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
230 'only_matching': True,
232 'url': 'http://www.bbc.co.uk/radio/player/p03cchwf',
233 'only_matching': True,
235 'url': 'https://www.bbc.co.uk/music/audiovideo/popular#p055bc55',
236 'only_matching': True,
238 'url': 'http://www.bbc.co.uk/programmes/w3csv1y9',
239 'only_matching': True,
242 _USP_RE = r'/([^/]+?)\.ism(?:\.hlsv2\.ism)?/[^/]+\.m3u8'
245 username, password = self._get_login_info()
249 login_page = self._download_webpage(
250 self._LOGIN_URL, None, 'Downloading signin page')
252 login_form = self._hidden_inputs(login_page)
255 'username': username,
256 'password': password,
259 post_url = urljoin(self._LOGIN_URL, self._search_regex(
260 r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
261 'post url', default=self._LOGIN_URL, group='url'))
263 response, urlh = self._download_webpage_handle(
264 post_url, None, 'Logging in', data=urlencode_postdata(login_form),
265 headers={'Referer': self._LOGIN_URL})
267 if self._LOGIN_URL in urlh.geturl():
268 error = clean_html(get_element_by_class('form-message', response))
270 raise ExtractorError(
271 'Unable to login: %s' % error, expected=True)
272 raise ExtractorError('Unable to log in')
274 def _real_initialize(self):
277 class MediaSelectionError(Exception):
278 def __init__(self, id):
281 def _extract_asx_playlist(self, connection, programme_id):
282 asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
283 return [ref.get('href') for ref in asx.findall('./Entry/ref')]
285 def _extract_items(self, playlist):
286 return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
288 def _findall_ns(self, element, xpath):
290 for ns in self._NAMESPACES:
291 elements.extend(element.findall(xpath % ns))
294 def _extract_medias(self, media_selection):
295 error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS)
297 media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS)
298 if error is not None:
299 raise BBCCoUkIE.MediaSelectionError(error.get('id'))
300 return self._findall_ns(media_selection, './{%s}media')
302 def _extract_connections(self, media):
303 return self._findall_ns(media, './{%s}connection')
305 def _get_subtitles(self, media, programme_id):
307 for connection in self._extract_connections(media):
308 captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
309 lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
312 'url': connection.get('href'),
318 def _raise_extractor_error(self, media_selection_error):
319 raise ExtractorError(
320 '%s returned error: %s' % (self.IE_NAME, media_selection_error.id),
323 def _download_media_selector(self, programme_id):
324 last_exception = None
325 for mediaselector_url in self._MEDIASELECTOR_URLS:
327 return self._download_media_selector_url(
328 mediaselector_url % programme_id, programme_id)
329 except BBCCoUkIE.MediaSelectionError as e:
330 if e.id in ('notukerror', 'geolocation', 'selectionunavailable'):
333 self._raise_extractor_error(e)
334 self._raise_extractor_error(last_exception)
336 def _download_media_selector_url(self, url, programme_id=None):
338 media_selection = self._download_xml(
339 url, programme_id, 'Downloading media selection XML')
340 except ExtractorError as ee:
341 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code in (403, 404):
342 media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8'))
345 return self._process_media_selector(media_selection, programme_id)
347 def _process_media_selector(self, media_selection, programme_id):
352 for media in self._extract_medias(media_selection):
353 kind = media.get('kind')
354 if kind in ('video', 'audio'):
355 bitrate = int_or_none(media.get('bitrate'))
356 encoding = media.get('encoding')
357 service = media.get('service')
358 width = int_or_none(media.get('width'))
359 height = int_or_none(media.get('height'))
360 file_size = int_or_none(media.get('media_file_size'))
361 for connection in self._extract_connections(media):
362 href = connection.get('href')
367 conn_kind = connection.get('kind')
368 protocol = connection.get('protocol')
369 supplier = connection.get('supplier')
370 transfer_format = connection.get('transferFormat')
371 format_id = supplier or conn_kind or protocol
373 format_id = '%s_%s' % (service, format_id)
375 if supplier == 'asx':
376 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
379 'format_id': 'ref%s_%s' % (i, format_id),
381 elif transfer_format == 'dash':
382 formats.extend(self._extract_mpd_formats(
383 href, programme_id, mpd_id=format_id, fatal=False))
384 elif transfer_format == 'hls':
385 formats.extend(self._extract_m3u8_formats(
386 href, programme_id, ext='mp4', entry_protocol='m3u8_native',
387 m3u8_id=format_id, fatal=False))
388 if re.search(self._USP_RE, href):
389 usp_formats = self._extract_m3u8_formats(
390 re.sub(self._USP_RE, r'/\1.ism/\1.m3u8', href),
391 programme_id, ext='mp4', entry_protocol='m3u8_native',
392 m3u8_id=format_id, fatal=False)
393 for f in usp_formats:
394 if f.get('height') and f['height'] > 720:
397 elif transfer_format == 'hds':
398 formats.extend(self._extract_f4m_formats(
399 href, programme_id, f4m_id=format_id, fatal=False))
401 if not service and not supplier and bitrate:
402 format_id += '-%d' % bitrate
404 'format_id': format_id,
405 'filesize': file_size,
420 if protocol in ('http', 'https'):
425 elif protocol == 'rtmp':
426 application = connection.get('application', 'ondemand')
427 auth_string = connection.get('authString')
428 identifier = connection.get('identifier')
429 server = connection.get('server')
431 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
432 'play_path': identifier,
433 'app': '%s?%s' % (application, auth_string),
434 'page_url': 'http://www.bbc.co.uk',
435 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
442 elif kind == 'captions':
443 subtitles = self.extract_subtitles(media, programme_id)
444 return formats, subtitles
446 def _download_playlist(self, playlist_id):
448 playlist = self._download_json(
449 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
450 playlist_id, 'Downloading playlist JSON')
452 version = playlist.get('defaultAvailableVersion')
454 smp_config = version['smpConfig']
455 title = smp_config['title']
456 description = smp_config['summary']
457 for item in smp_config['items']:
459 if kind not in ('programme', 'radioProgramme'):
461 programme_id = item.get('vpid')
462 duration = int_or_none(item.get('duration'))
463 formats, subtitles = self._download_media_selector(programme_id)
464 return programme_id, title, description, duration, formats, subtitles
465 except ExtractorError as ee:
466 if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
469 # fallback to legacy playlist
470 return self._process_legacy_playlist(playlist_id)
472 def _process_legacy_playlist_url(self, url, display_id):
473 playlist = self._download_legacy_playlist_url(url, display_id)
474 return self._extract_from_legacy_playlist(playlist, display_id)
476 def _process_legacy_playlist(self, playlist_id):
477 return self._process_legacy_playlist_url(
478 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
480 def _download_legacy_playlist_url(self, url, playlist_id=None):
481 return self._download_xml(
482 url, playlist_id, 'Downloading legacy playlist XML')
484 def _extract_from_legacy_playlist(self, playlist, playlist_id):
485 no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS)
486 if no_items is not None:
487 reason = no_items.get('reason')
488 if reason == 'preAvailability':
489 msg = 'Episode %s is not yet available' % playlist_id
490 elif reason == 'postAvailability':
491 msg = 'Episode %s is no longer available' % playlist_id
492 elif reason == 'noMedia':
493 msg = 'Episode %s is not currently available' % playlist_id
495 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
496 raise ExtractorError(msg, expected=True)
498 for item in self._extract_items(playlist):
499 kind = item.get('kind')
500 if kind not in ('programme', 'radioProgramme'):
502 title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
503 description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
504 description = description_el.text if description_el is not None else None
506 def get_programme_id(item):
507 def get_from_attributes(item):
508 for p in('identifier', 'group'):
510 if value and re.match(r'^[pb][\da-z]{7}$', value):
512 get_from_attributes(item)
513 mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS)
514 if mediator is not None:
515 return get_from_attributes(mediator)
517 programme_id = get_programme_id(item)
518 duration = int_or_none(item.get('duration'))
521 formats, subtitles = self._download_media_selector(programme_id)
523 formats, subtitles = self._process_media_selector(item, playlist_id)
524 programme_id = playlist_id
526 return programme_id, title, description, duration, formats, subtitles
528 def _real_extract(self, url):
529 group_id = self._match_id(url)
531 webpage = self._download_webpage(url, group_id, 'Downloading video page')
533 error = self._search_regex(
534 r'<div\b[^>]+\bclass=["\']smp__message delta["\'][^>]*>([^<]+)<',
535 webpage, 'error', default=None)
537 raise ExtractorError(error, expected=True)
542 tviplayer = self._search_regex(
543 r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
544 webpage, 'player', default=None)
547 player = self._parse_json(tviplayer, group_id).get('player', {})
548 duration = int_or_none(player.get('duration'))
549 programme_id = player.get('vpid')
552 programme_id = self._search_regex(
553 r'"vpid"\s*:\s*"(%s)"' % self._ID_REGEX, webpage, 'vpid', fatal=False, default=None)
556 formats, subtitles = self._download_media_selector(programme_id)
557 title = self._og_search_title(webpage, default=None) or self._html_search_regex(
558 (r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>',
559 r'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage, 'title')
560 description = self._search_regex(
561 (r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
562 r'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'),
563 webpage, 'description', default=None)
565 description = self._html_search_meta('description', webpage)
567 programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
569 self._sort_formats(formats)
574 'description': description,
575 'thumbnail': self._og_search_thumbnail(webpage, default=None),
576 'duration': duration,
578 'subtitles': subtitles,
582 class BBCIE(BBCCoUkIE):
585 _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
587 _MEDIASELECTOR_URLS = [
588 # Provides HQ HLS streams but fails with geolocation in some cases when it's
589 # even not geo restricted at all
590 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
591 # Provides more formats, namely direct mp4 links, but fails on some videos with
592 # notukerror for non UK (?) users (e.g.
593 # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
594 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
595 # Provides fewer formats, but works everywhere for everybody (hopefully)
596 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
600 # article with multiple videos embedded with data-playable containing vpids
601 'url': 'http://www.bbc.com/news/world-europe-32668511',
603 'id': 'world-europe-32668511',
604 'title': 'Russia stages massive WW2 parade despite Western boycott',
605 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
609 # article with multiple videos embedded with data-playable (more videos)
610 'url': 'http://www.bbc.com/news/business-28299555',
612 'id': 'business-28299555',
613 'title': 'Farnborough Airshow: Video highlights',
614 'description': 'BBC reports and video highlights at the Farnborough Airshow.',
619 # article with multiple videos embedded with `new SMP()`
621 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
623 'id': '3662a707-0af9-3149-963f-47bea720b460',
626 'playlist_count': 18,
628 # single video embedded with data-playable containing vpid
629 'url': 'http://www.bbc.com/news/world-europe-32041533',
633 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
634 'description': 'md5:2868290467291b37feda7863f7a83f54',
636 'timestamp': 1427219242,
637 'upload_date': '20150324',
641 'skip_download': True,
644 # article with single video embedded with data-playable containing XML playlist
645 # with direct video links as progressiveDownloadUrl (for now these are extracted)
646 # and playlist with f4m and m3u8 as streamingUrl
647 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
649 'id': '150615_telabyad_kentin_cogu',
651 'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
652 'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
653 'timestamp': 1434397334,
654 'upload_date': '20150615',
657 'skip_download': True,
660 # single video embedded with data-playable containing XML playlists (regional section)
661 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
663 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
665 'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
666 'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
667 'timestamp': 1434713142,
668 'upload_date': '20150619',
671 'skip_download': True,
674 # single video from video playlist embedded with vxp-playlist-data JSON
675 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
679 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
681 'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
684 'skip_download': True,
687 # single video story with digitalData
688 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
692 'title': 'Sri Lanka’s spicy secret',
693 'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
694 'timestamp': 1437674293,
695 'upload_date': '20150723',
699 'skip_download': True,
702 # single video story without digitalData
703 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
707 'title': 'Hyundai Santa Fe Sport: Rock star',
708 'description': 'md5:b042a26142c4154a6e472933cf20793d',
709 'timestamp': 1415867444,
710 'upload_date': '20141113',
714 'skip_download': True,
717 # single video embedded with Morph
718 'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
722 'title': "Nigeria v Japan - Men's First Round",
723 'description': 'Live coverage of the first round from Group B at the Amazonia Arena.',
725 'uploader': 'BBC Sport',
726 'uploader_id': 'bbc_sport',
730 'skip_download': True,
732 'skip': 'Georestricted to UK',
734 # single video with playlist.sxml URL in playlist param
735 'url': 'http://www.bbc.com/sport/0/football/33653409',
739 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
740 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
745 'skip_download': True,
748 # article with multiple videos embedded with playlist.sxml in playlist param
749 'url': 'http://www.bbc.com/sport/0/football/34475836',
752 'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
753 'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
757 # school report article with single video
758 'url': 'http://www.bbc.co.uk/schoolreport/35744779',
761 'title': 'School which breaks down barriers in Jerusalem',
765 # single video with playlist URL from weather section
766 'url': 'http://www.bbc.com/weather/features/33601775',
767 'only_matching': True,
769 # custom redirection to www.bbc.com
770 'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
771 'only_matching': True,
773 # single video article embedded with data-media-vpid
774 'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
775 'only_matching': True,
777 'url': 'https://www.bbc.co.uk/bbcthree/clip/73d0bbd0-abc3-4cea-b3c0-cdae21905eb1',
781 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
782 'description': 'md5:4b7dfd063d5a789a1512e99662be3ddd',
785 'skip_download': True,
790 def suitable(cls, url):
791 EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE)
792 return (False if any(ie.suitable(url) for ie in EXCLUDE_IE)
793 else super(BBCIE, cls).suitable(url))
795 def _extract_from_media_meta(self, media_meta, video_id):
796 # Direct links to media in media metadata (e.g.
797 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
798 # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
799 source_files = media_meta.get('sourceFiles')
803 'format_id': format_id,
804 'ext': f.get('encoding'),
805 'tbr': float_or_none(f.get('bitrate'), 1000),
806 'filesize': int_or_none(f.get('filesize')),
807 } for format_id, f in source_files.items() if f.get('url')], []
809 programme_id = media_meta.get('externalId')
811 return self._download_media_selector(programme_id)
813 # Process playlist.sxml as legacy playlist
814 href = media_meta.get('href')
816 playlist = self._download_legacy_playlist_url(href)
817 _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
818 return formats, subtitles
822 def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
823 programme_id, title, description, duration, formats, subtitles = \
824 self._process_legacy_playlist_url(url, playlist_id)
825 self._sort_formats(formats)
829 'description': description,
830 'duration': duration,
831 'timestamp': timestamp,
833 'subtitles': subtitles,
836 def _real_extract(self, url):
837 playlist_id = self._match_id(url)
839 webpage = self._download_webpage(url, playlist_id)
841 json_ld_info = self._search_json_ld(webpage, playlist_id, default={})
842 timestamp = json_ld_info.get('timestamp')
844 playlist_title = json_ld_info.get('title')
845 if not playlist_title:
846 playlist_title = self._og_search_title(
847 webpage, default=None) or self._html_search_regex(
848 r'<title>(.+?)</title>', webpage, 'playlist title', default=None)
850 playlist_title = re.sub(r'(.+)\s*-\s*BBC.*?$', r'\1', playlist_title).strip()
852 playlist_description = json_ld_info.get(
853 'description') or self._og_search_description(webpage, default=None)
856 timestamp = parse_iso8601(self._search_regex(
857 [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
858 r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
859 r'"datePublished":\s*"([^"]+)'],
860 webpage, 'date', default=None))
864 # article with multiple videos embedded with playlist.sxml (e.g.
865 # http://www.bbc.com/sport/0/football/34475836)
866 playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
867 playlists.extend(re.findall(r'data-media-id="([^"]+/playlist\.sxml)"', webpage))
870 self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
871 for playlist_url in playlists]
873 # news article with multiple videos embedded with data-playable
874 data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
876 for _, data_playable_json in data_playables:
877 data_playable = self._parse_json(
878 unescapeHTML(data_playable_json), playlist_id, fatal=False)
879 if not data_playable:
881 settings = data_playable.get('settings', {})
883 # data-playable with video vpid in settings.playlistObject.items (e.g.
884 # http://www.bbc.com/news/world-us-canada-34473351)
885 playlist_object = settings.get('playlistObject', {})
887 items = playlist_object.get('items')
888 if items and isinstance(items, list):
889 title = playlist_object['title']
890 description = playlist_object.get('summary')
891 duration = int_or_none(items[0].get('duration'))
892 programme_id = items[0].get('vpid')
893 formats, subtitles = self._download_media_selector(programme_id)
894 self._sort_formats(formats)
898 'description': description,
899 'timestamp': timestamp,
900 'duration': duration,
902 'subtitles': subtitles,
905 # data-playable without vpid but with a playlist.sxml URLs
906 # in otherSettings.playlist (e.g.
907 # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
908 playlist = data_playable.get('otherSettings', {}).get('playlist', {})
911 for key in ('streaming', 'progressiveDownload'):
912 playlist_url = playlist.get('%sUrl' % key)
916 info = self._extract_from_playlist_sxml(
917 playlist_url, playlist_id, timestamp)
921 entry['title'] = info['title']
922 entry['formats'].extend(info['formats'])
923 except Exception as e:
924 # Some playlist URL may fail with 500, at the same time
925 # the other one may work fine (e.g.
926 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
927 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
931 self._sort_formats(entry['formats'])
932 entries.append(entry)
935 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
937 # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
938 programme_id = self._search_regex(
939 [r'data-(?:video-player|media)-vpid="(%s)"' % self._ID_REGEX,
940 r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
941 r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
942 webpage, 'vpid', default=None)
945 formats, subtitles = self._download_media_selector(programme_id)
946 self._sort_formats(formats)
947 # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
948 digital_data = self._parse_json(
950 r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
951 programme_id, fatal=False)
952 page_info = digital_data.get('page', {}).get('pageInfo', {})
953 title = page_info.get('pageName') or self._og_search_title(webpage)
954 description = page_info.get('description') or self._og_search_description(webpage)
955 timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
959 'description': description,
960 'timestamp': timestamp,
962 'subtitles': subtitles,
965 # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
966 # There are several setPayload calls may be present but the video
967 # seems to be always related to the first one
968 morph_payload = self._parse_json(
970 r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
971 webpage, 'morph payload', default='{}'),
972 playlist_id, fatal=False)
974 components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
975 for component in components:
976 if not isinstance(component, dict):
978 lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
981 identifiers = lead_media.get('identifiers')
982 if not identifiers or not isinstance(identifiers, dict):
984 programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
987 title = lead_media.get('title') or self._og_search_title(webpage)
988 formats, subtitles = self._download_media_selector(programme_id)
989 self._sort_formats(formats)
990 description = lead_media.get('summary')
991 uploader = lead_media.get('masterBrand')
992 uploader_id = lead_media.get('mid')
994 duration_d = lead_media.get('duration')
995 if isinstance(duration_d, dict):
996 duration = parse_duration(dict_get(
997 duration_d, ('rawDuration', 'formattedDuration', 'spokenDuration')))
1001 'description': description,
1002 'duration': duration,
1003 'uploader': uploader,
1004 'uploader_id': uploader_id,
1006 'subtitles': subtitles,
1009 bbc3_config = self._parse_json(
1011 r'(?s)bbcthreeConfig\s*=\s*({.+?})\s*;\s*<', webpage,
1012 'bbcthree config', default='{}'),
1013 playlist_id, transform_source=js_to_json, fatal=False)
1015 bbc3_playlist = try_get(
1016 bbc3_config, lambda x: x['payload']['content']['bbcMedia']['playlist'],
1019 playlist_title = bbc3_playlist.get('title') or playlist_title
1020 thumbnail = bbc3_playlist.get('holdingImageURL')
1022 for bbc3_item in bbc3_playlist['items']:
1023 programme_id = bbc3_item.get('versionID')
1024 if not programme_id:
1026 formats, subtitles = self._download_media_selector(programme_id)
1027 self._sort_formats(formats)
1030 'title': playlist_title,
1031 'thumbnail': thumbnail,
1032 'timestamp': timestamp,
1034 'subtitles': subtitles,
1036 return self.playlist_result(
1037 entries, playlist_id, playlist_title, playlist_description)
1039 def extract_all(pattern):
1040 return list(filter(None, map(
1041 lambda s: self._parse_json(s, playlist_id, fatal=False),
1042 re.findall(pattern, webpage))))
1044 # Multiple video article (e.g.
1045 # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
1046 EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX
1048 for match in extract_all(r'new\s+SMP\(({.+?})\)'):
1049 embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
1050 if embed_url and re.match(EMBED_URL, embed_url):
1051 entries.append(embed_url)
1052 entries.extend(re.findall(
1053 r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
1055 return self.playlist_result(
1056 [self.url_result(entry_, 'BBCCoUk') for entry_ in entries],
1057 playlist_id, playlist_title, playlist_description)
1059 # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
1060 medias = extract_all(r"data-media-meta='({[^']+})'")
1063 # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
1064 media_asset = self._search_regex(
1065 r'mediaAssetPage\.init\(\s*({.+?}), "/',
1066 webpage, 'media asset', default=None)
1068 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
1070 for video in media_asset_page.get('videos', {}).values():
1071 medias.extend(video.values())
1074 # Multiple video playlist with single `now playing` entry (e.g.
1075 # http://www.bbc.com/news/video_and_audio/must_see/33767813)
1076 vxp_playlist = self._parse_json(
1078 r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>',
1079 webpage, 'playlist data'),
1081 playlist_medias = []
1082 for item in vxp_playlist:
1083 media = item.get('media')
1086 playlist_medias.append(media)
1087 # Download single video if found media with asset id matching the video id from URL
1088 if item.get('advert', {}).get('assetId') == playlist_id:
1091 # Fallback to the whole playlist
1093 medias = playlist_medias
1096 for num, media_meta in enumerate(medias, start=1):
1097 formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
1100 self._sort_formats(formats)
1102 video_id = media_meta.get('externalId')
1104 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
1106 title = media_meta.get('caption')
1108 title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
1110 duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
1113 for image in media_meta.get('images', {}).values():
1114 images.extend(image.values())
1115 if 'image' in media_meta:
1116 images.append(media_meta['image'])
1119 'url': image.get('href'),
1120 'width': int_or_none(image.get('width')),
1121 'height': int_or_none(image.get('height')),
1122 } for image in images]
1127 'thumbnails': thumbnails,
1128 'duration': duration,
1129 'timestamp': timestamp,
1131 'subtitles': subtitles,
1134 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
1137 class BBCCoUkArticleIE(InfoExtractor):
1138 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
1139 IE_NAME = 'bbc.co.uk:article'
1140 IE_DESC = 'BBC articles'
1143 'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer',
1145 'id': '3jNQLTMrPlYGTBn0WV6M2MS',
1146 'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four',
1147 'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.',
1149 'playlist_count': 4,
1150 'add_ie': ['BBCCoUk'],
1153 def _real_extract(self, url):
1154 playlist_id = self._match_id(url)
1156 webpage = self._download_webpage(url, playlist_id)
1158 title = self._og_search_title(webpage)
1159 description = self._og_search_description(webpage).strip()
1161 entries = [self.url_result(programme_url) for programme_url in re.findall(
1162 r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
1164 return self.playlist_result(entries, playlist_id, title, description)
1167 class BBCCoUkPlaylistBaseIE(InfoExtractor):
1168 def _entries(self, webpage, url, playlist_id):
1169 single_page = 'page' in compat_urlparse.parse_qs(
1170 compat_urlparse.urlparse(url).query)
1171 for page_num in itertools.count(2):
1172 for video_id in re.findall(
1173 self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage):
1174 yield self.url_result(
1175 self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
1178 next_page = self._search_regex(
1179 r'<li[^>]+class=(["\'])pagination_+next\1[^>]*><a[^>]+href=(["\'])(?P<url>(?:(?!\2).)+)\2',
1180 webpage, 'next page url', default=None, group='url')
1183 webpage = self._download_webpage(
1184 compat_urlparse.urljoin(url, next_page), playlist_id,
1185 'Downloading page %d' % page_num, page_num)
1187 def _real_extract(self, url):
1188 playlist_id = self._match_id(url)
1190 webpage = self._download_webpage(url, playlist_id)
1192 title, description = self._extract_title_and_description(webpage)
1194 return self.playlist_result(
1195 self._entries(webpage, url, playlist_id),
1196 playlist_id, title, description)
1199 class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
1200 IE_NAME = 'bbc.co.uk:iplayer:playlist'
1201 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
1202 _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
1203 _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
1205 'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v',
1208 'title': 'The Disappearance',
1209 'description': 'French thriller serial about a missing teenager.',
1211 'playlist_mincount': 6,
1212 'skip': 'This programme is not currently available on BBC iPlayer',
1214 # Available for over a year unlike 30 days for most other programmes
1215 'url': 'http://www.bbc.co.uk/iplayer/group/p02tcc32',
1218 'title': 'Bohemian Icons',
1219 'description': 'md5:683e901041b2fe9ba596f2ab04c4dbe7',
1221 'playlist_mincount': 10,
1224 def _extract_title_and_description(self, webpage):
1225 title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
1226 description = self._search_regex(
1227 r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
1228 webpage, 'description', fatal=False, group='value')
1229 return title, description
1232 class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
1233 IE_NAME = 'bbc.co.uk:playlist'
1234 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
1235 _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
1236 _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
1238 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1241 'title': 'The Disappearance - Clips - BBC Four',
1242 'description': 'French thriller serial about a missing teenager.',
1244 'playlist_mincount': 7,
1246 # multipage playlist, explicit page
1247 'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips?page=1',
1250 'title': 'Frozen Planet - Clips - BBC One',
1251 'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1253 'playlist_mincount': 24,
1255 # multipage playlist, all pages
1256 'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips',
1259 'title': 'Frozen Planet - Clips - BBC One',
1260 'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1262 'playlist_mincount': 142,
1264 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/broadcasts/2016/06',
1265 'only_matching': True,
1267 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1268 'only_matching': True,
1270 'url': 'http://www.bbc.co.uk/programmes/b055jkys/episodes/player',
1271 'only_matching': True,
1274 def _extract_title_and_description(self, webpage):
1275 title = self._og_search_title(webpage, fatal=False)
1276 description = self._og_search_description(webpage)
1277 return title, description