2 from __future__ import unicode_literals
5 import xml.etree.ElementTree
7 from .common import InfoExtractor
15 from ..compat import compat_HTTPError
18 class BBCCoUkIE(InfoExtractor):
20 IE_DESC = 'BBC iPlayer'
21 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
23 _MEDIASELECTOR_URL = 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s'
27 'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
31 'title': 'Kaleidoscope, Leonard Cohen',
32 'description': 'The Canadian poet and songwriter reflects on his musical career.',
37 'skip_download': True,
41 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
45 'title': 'The Man in Black: Series 3: The Printed Name',
46 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
51 'skip_download': True,
53 'skip': 'Episode is no longer available on BBC iPlayer Radio',
56 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
60 'title': 'The Voice UK: Series 3: Blind Auditions 5',
61 '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.",
66 'skip_download': True,
68 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
71 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
75 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
76 'description': '2. Invasion',
81 'skip_download': True,
83 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
85 'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
89 'title': 'Pete Tong, The Essential New Tune Special',
90 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
95 'skip_download': True,
98 'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
103 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
104 'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
109 'skip_download': True,
112 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
117 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
118 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
123 'skip_download': True,
126 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
130 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
131 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
136 'skip_download': True,
138 'skip': 'geolocation',
140 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
144 '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.',
145 'title': 'Royal Academy Summer Exhibition',
150 'skip_download': True,
152 'skip': 'geolocation',
154 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
155 'only_matching': True,
157 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
158 'only_matching': True,
160 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
161 'only_matching': True,
165 def _extract_asx_playlist(self, connection, programme_id):
166 asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
167 return [ref.get('href') for ref in asx.findall('./Entry/ref')]
169 def _extract_connection(self, connection, programme_id):
171 protocol = connection.get('protocol')
172 supplier = connection.get('supplier')
173 if protocol == 'http':
174 href = connection.get('href')
176 if supplier == 'asx':
177 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
180 'format_id': 'ref%s_%s' % (i, supplier),
186 'format_id': supplier,
188 elif protocol == 'rtmp':
189 application = connection.get('application', 'ondemand')
190 auth_string = connection.get('authString')
191 identifier = connection.get('identifier')
192 server = connection.get('server')
194 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
195 'play_path': identifier,
196 'app': '%s?%s' % (application, auth_string),
197 'page_url': 'http://www.bbc.co.uk',
198 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
201 'format_id': supplier,
205 def _extract_items(self, playlist):
206 return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
208 def _extract_medias(self, media_selection):
209 error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
210 if error is not None:
211 raise ExtractorError(
212 '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
213 return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
215 def _extract_connections(self, media):
216 return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
218 def _extract_video(self, media, programme_id):
220 vbr = int_or_none(media.get('bitrate'))
221 vcodec = media.get('encoding')
222 service = media.get('service')
223 width = int_or_none(media.get('width'))
224 height = int_or_none(media.get('height'))
225 file_size = int_or_none(media.get('media_file_size'))
226 for connection in self._extract_connections(media):
227 conn_formats = self._extract_connection(connection, programme_id)
228 for format in conn_formats:
230 'format_id': '%s_%s' % (service, format['format_id']),
235 'filesize': file_size,
237 formats.extend(conn_formats)
240 def _extract_audio(self, media, programme_id):
242 abr = int_or_none(media.get('bitrate'))
243 acodec = media.get('encoding')
244 service = media.get('service')
245 for connection in self._extract_connections(media):
246 conn_formats = self._extract_connection(connection, programme_id)
247 for format in conn_formats:
249 'format_id': '%s_%s' % (service, format['format_id']),
253 formats.extend(conn_formats)
256 def _get_subtitles(self, media, programme_id):
258 for connection in self._extract_connections(media):
259 captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
260 lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
263 'url': connection.get('href'),
269 def _download_media_selector(self, programme_id):
270 return self._download_media_selector_url(
271 self._MEDIASELECTOR_URL % programme_id, programme_id)
273 def _download_media_selector_url(self, url, programme_id=None):
275 media_selection = self._download_xml(
276 url, programme_id, 'Downloading media selection XML')
277 except ExtractorError as ee:
278 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
279 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8'))
282 return self._process_media_selector(media_selection, programme_id)
284 def _process_media_selector(self, media_selection, programme_id):
288 for media in self._extract_medias(media_selection):
289 kind = media.get('kind')
291 formats.extend(self._extract_audio(media, programme_id))
292 elif kind == 'video':
293 formats.extend(self._extract_video(media, programme_id))
294 elif kind == 'captions':
295 subtitles = self.extract_subtitles(media, programme_id)
297 return formats, subtitles
299 def _download_playlist(self, playlist_id):
301 playlist = self._download_json(
302 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
303 playlist_id, 'Downloading playlist JSON')
305 version = playlist.get('defaultAvailableVersion')
307 smp_config = version['smpConfig']
308 title = smp_config['title']
309 description = smp_config['summary']
310 for item in smp_config['items']:
312 if kind != 'programme' and kind != 'radioProgramme':
314 programme_id = item.get('vpid')
315 duration = int_or_none(item.get('duration'))
316 formats, subtitles = self._download_media_selector(programme_id)
317 return programme_id, title, description, duration, formats, subtitles
318 except ExtractorError as ee:
319 if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
322 # fallback to legacy playlist
323 return self._process_legacy_playlist(playlist_id)
325 def _process_legacy_playlist_url(self, url, display_id):
326 playlist = self._download_legacy_playlist_url(url, display_id)
327 return self._extract_from_legacy_playlist(playlist, display_id)
329 def _process_legacy_playlist(self, playlist_id):
330 return self._process_legacy_playlist_url(
331 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
333 def _download_legacy_playlist_url(self, url, playlist_id=None):
334 return self._download_xml(
335 url, playlist_id, 'Downloading legacy playlist XML')
337 def _extract_from_legacy_playlist(self, playlist, playlist_id):
338 no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
339 if no_items is not None:
340 reason = no_items.get('reason')
341 if reason == 'preAvailability':
342 msg = 'Episode %s is not yet available' % playlist_id
343 elif reason == 'postAvailability':
344 msg = 'Episode %s is no longer available' % playlist_id
345 elif reason == 'noMedia':
346 msg = 'Episode %s is not currently available' % playlist_id
348 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
349 raise ExtractorError(msg, expected=True)
351 for item in self._extract_items(playlist):
352 kind = item.get('kind')
353 if kind != 'programme' and kind != 'radioProgramme':
355 title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
356 description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
358 def get_programme_id(item):
359 def get_from_attributes(item):
360 for p in('identifier', 'group'):
362 if value and re.match(r'^[pb][\da-z]{7}$', value):
364 get_from_attributes(item)
365 mediator = item.find('./{http://bbc.co.uk/2008/emp/playlist}mediator')
366 if mediator is not None:
367 return get_from_attributes(mediator)
369 programme_id = get_programme_id(item)
370 duration = int_or_none(item.get('duration'))
371 # TODO: programme_id can be None and media items can be incorporated right inside
372 # playlist's item (e.g. http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
374 formats, subtitles = self._download_media_selector(programme_id)
376 return programme_id, title, description, duration, formats, subtitles
378 def _real_extract(self, url):
379 group_id = self._match_id(url)
381 webpage = self._download_webpage(url, group_id, 'Downloading video page')
385 tviplayer = self._search_regex(
386 r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
387 webpage, 'player', default=None)
390 player = self._parse_json(tviplayer, group_id).get('player', {})
391 duration = int_or_none(player.get('duration'))
392 programme_id = player.get('vpid')
395 programme_id = self._search_regex(
396 r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
399 formats, subtitles = self._download_media_selector(programme_id)
400 title = self._og_search_title(webpage)
401 description = self._search_regex(
402 r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
403 webpage, 'description', fatal=False)
405 programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
407 self._sort_formats(formats)
412 'description': description,
413 'thumbnail': self._og_search_thumbnail(webpage, default=None),
414 'duration': duration,
416 'subtitles': subtitles,
420 class BBCIE(BBCCoUkIE):
423 _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
425 # fails with notukerror for some videos
426 #_MEDIASELECTOR_URL = 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s'
427 _MEDIASELECTOR_URL = 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s'
430 # article with multiple videos embedded with data-media-meta containing
431 # playlist.sxml, externalId and no direct video links
432 'url': 'http://www.bbc.com/news/world-europe-32668511',
434 'id': 'world-europe-32668511',
435 'title': 'Russia stages massive WW2 parade despite Western boycott',
436 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
440 # article with multiple videos embedded with data-media-meta (more videos)
441 'url': 'http://www.bbc.com/news/business-28299555',
443 'id': 'business-28299555',
444 'title': 'Farnborough Airshow: Video highlights',
445 'description': 'BBC reports and video highlights at the Farnborough Airshow.',
450 # single video embedded with mediaAssetPage.init()
451 'url': 'http://www.bbc.com/news/world-europe-32041533',
455 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
457 'timestamp': 1427219242,
458 'upload_date': '20150324',
462 'skip_download': True,
465 # article with single video embedded with data-media-meta containing
466 # direct video links (for now these are extracted) and playlist.xml (with
467 # media items as f4m and m3u8 - currently unsupported)
468 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
470 'id': '150615_telabyad_kentin_cogu',
472 'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
474 'timestamp': 1434397334,
475 'upload_date': '20150615',
478 'skip_download': True,
481 # single video embedded with mediaAssetPage.init() (regional section)
482 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
484 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
486 'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
488 'timestamp': 1434713142,
489 'upload_date': '20150619',
492 'skip_download': True,
495 # single video story with digitalData
496 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
500 'title': 'Sri Lanka’s spicy secret',
501 'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
502 'timestamp': 1437674293,
503 'upload_date': '20150723',
507 'skip_download': True,
510 # single video story without digitalData
511 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
515 'title': 'Hyundai Santa Fe Sport: Rock star',
516 'description': 'md5:b042a26142c4154a6e472933cf20793d',
517 'timestamp': 1368473503,
518 'upload_date': '20130513',
522 'skip_download': True,
525 # single video with playlist.sxml URL
526 'url': 'http://www.bbc.com/sport/0/football/33653409',
530 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
531 'description': 'md5:398fca0e2e701c609d726e034fa1fc89',
536 'skip_download': True,
539 # single video with playlist URL from weather section
540 'url': 'http://www.bbc.com/weather/features/33601775',
541 'only_matching': True,
543 # custom redirection to www.bbc.com
544 'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
545 'only_matching': True,
549 def suitable(cls, url):
550 return False if BBCCoUkIE.suitable(url) else super(BBCIE, cls).suitable(url)
552 def _extract_from_media_meta(self, media_meta, video_id):
553 # Direct links to media in media metadata (e.g.
554 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
555 # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
556 source_files = media_meta.get('sourceFiles')
560 'format_id': format_id,
561 'ext': f.get('encoding'),
562 'tbr': float_or_none(f.get('bitrate'), 1000),
563 'filesize': int_or_none(f.get('filesize')),
564 } for format_id, f in source_files.items() if f.get('url')], []
566 programme_id = media_meta.get('externalId')
568 return self._download_media_selector(programme_id)
570 # Process playlist.sxml as legacy playlist
571 href = media_meta.get('href')
573 playlist = self._download_legacy_playlist_url(href)
574 _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
575 return formats, subtitles
579 def _real_extract(self, url):
580 playlist_id = self._match_id(url)
582 webpage = self._download_webpage(url, playlist_id)
584 timestamp = parse_iso8601(self._search_regex(
585 [r'"datePublished":\s*"([^"]+)',
586 r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
587 r'itemprop="datePublished"[^>]+datetime="([^"]+)"'],
588 webpage, 'date', default=None))
590 # single video with playlist.sxml URL (e.g. http://www.bbc.com/sport/0/football/3365340ng)
591 playlist = self._search_regex(
592 r'<param[^>]+name="playlist"[^>]+value="([^"]+)"',
593 webpage, 'playlist', default=None)
595 programme_id, title, description, duration, formats, subtitles = \
596 self._process_legacy_playlist_url(playlist, playlist_id)
597 self._sort_formats(formats)
601 'description': description,
602 'duration': duration,
603 'timestamp': timestamp,
605 'subtitles': subtitles,
608 # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
609 programme_id = self._search_regex(
610 [r'data-video-player-vpid="([\da-z]{8})"',
611 r'<param[^>]+name="externalIdentifier"[^>]+value="([\da-z]{8})"'],
612 webpage, 'vpid', default=None)
614 formats, subtitles = self._download_media_selector(programme_id)
615 self._sort_formats(formats)
616 # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
617 digital_data = self._parse_json(
619 r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
620 programme_id, fatal=False)
621 page_info = digital_data.get('page', {}).get('pageInfo', {})
622 title = page_info.get('pageName') or self._og_search_title(webpage)
623 description = page_info.get('description') or self._og_search_description(webpage)
624 timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
628 'description': description,
629 'timestamp': timestamp,
631 'subtitles': subtitles,
634 playlist_title = self._html_search_regex(
635 r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'playlist title')
636 playlist_description = self._og_search_description(webpage)
638 # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
639 medias = list(filter(None, map(
640 lambda s: self._parse_json(s, playlist_id, fatal=False),
641 re.findall(r"data-media-meta='({[^']+})'", webpage))))
644 # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
645 media_asset_page = self._parse_json(
647 r'mediaAssetPage\.init\(\s*({.+?}), "/', webpage, 'media asset'),
650 for video in media_asset_page.get('videos', {}).values():
651 medias.extend(video.values())
654 for num, media_meta in enumerate(medias, start=1):
655 formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
658 self._sort_formats(formats)
660 video_id = media_meta.get('externalId')
662 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
664 title = media_meta.get('caption')
666 title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
668 duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
671 for image in media_meta.get('images', {}).values():
672 images.extend(image.values())
673 if 'image' in media_meta:
674 images.append(media_meta['image'])
677 'url': image.get('href'),
678 'width': int_or_none(image.get('width')),
679 'height': int_or_none(image.get('height')),
680 } for image in images]
685 'thumbnails': thumbnails,
686 'duration': duration,
687 'timestamp': timestamp,
689 'subtitles': subtitles,
692 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)