2 from __future__
import unicode_literals
6 from .common
import InfoExtractor
15 from ..compat
import (
16 compat_etree_fromstring
,
21 class BBCCoUkIE(InfoExtractor
):
23 IE_DESC
= 'BBC iPlayer'
24 _ID_REGEX
= r
'[pb][\da-z]{7}'
27 (?:www\.)?bbc\.co\.uk/
29 programmes/(?!articles/)|
30 iplayer(?:/[^/]+)?/(?:episode/|playlist/)|
37 _MEDIASELECTOR_URLS
= [
38 # Provides HQ HLS streams with even better quality that pc mediaset but fails
39 # with geolocation in some cases when it's even not geo restricted at all (e.g.
40 # http://www.bbc.co.uk/programmes/b06bp7lf). Also may fail with selectionunavailable.
41 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
42 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
45 _MEDIASELECTION_NS
= 'http://bbc.co.uk/2008/mp/mediaselection'
46 _EMP_PLAYLIST_NS
= 'http://bbc.co.uk/2008/emp/playlist'
55 'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
59 'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
60 'description': 'The Canadian poet and songwriter reflects on his musical career.',
64 'skip_download': True,
68 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
72 'title': 'The Man in Black: Series 3: The Printed Name',
73 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
78 'skip_download': True,
80 'skip': 'Episode is no longer available on BBC iPlayer Radio',
83 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
87 'title': 'The Voice UK: Series 3: Blind Auditions 5',
88 '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.',
93 'skip_download': True,
95 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
98 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
102 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
103 'description': '2. Invasion',
108 'skip_download': True,
110 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
112 'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
116 'title': 'Pete Tong, The Essential New Tune Special',
117 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
122 'skip_download': True,
124 'skip': 'Episode is no longer available on BBC iPlayer Radio',
126 'url': 'http://www.bbc.co.uk/music/clips/p022h44b',
131 'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
132 'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
137 'skip_download': True,
140 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
145 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
146 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
151 'skip_download': True,
154 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
158 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
159 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
164 'skip_download': True,
166 'skip': 'geolocation',
168 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
172 '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.',
173 'title': 'Royal Academy Summer Exhibition',
178 'skip_download': True,
180 'skip': 'geolocation',
182 # iptv-all mediaset fails with geolocation however there is no geo restriction
183 # for this programme at all
184 'url': 'http://www.bbc.co.uk/programmes/b06rkn85',
188 'title': "Best of the Mini-Mixes 2015: Part 3, Annie Mac's Friday Night - BBC Radio 1",
189 'description': "Annie has part three in the Best of the Mini-Mixes 2015, plus the year's Most Played!",
193 'skip_download': True,
196 # compact player (https://github.com/rg3/youtube-dl/issues/8147)
197 'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
201 'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
202 'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
206 'skip_download': True,
209 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
210 'only_matching': True,
212 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
213 'only_matching': True,
215 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
216 'only_matching': True,
218 'url': 'http://www.bbc.co.uk/radio/player/p03cchwf',
219 'only_matching': True,
223 class MediaSelectionError(Exception):
224 def __init__(self
, id):
227 def _extract_asx_playlist(self
, connection
, programme_id
):
228 asx
= self
._download
_xml
(connection
.get('href'), programme_id
, 'Downloading ASX playlist')
229 return [ref
.get('href') for ref
in asx
.findall('./Entry/ref')]
231 def _extract_connection(self
, connection
, programme_id
):
233 kind
= connection
.get('kind')
234 protocol
= connection
.get('protocol')
235 supplier
= connection
.get('supplier')
236 if protocol
== 'http':
237 href
= connection
.get('href')
238 transfer_format
= connection
.get('transferFormat')
240 if supplier
== 'asx':
241 for i
, ref
in enumerate(self
._extract
_asx
_playlist
(connection
, programme_id
)):
244 'format_id': 'ref%s_%s' % (i
, supplier
),
246 # Skip DASH until supported
247 elif transfer_format
== 'dash':
249 elif transfer_format
== 'hls':
250 formats
.extend(self
._extract
_m
3u8_formats
(
251 href
, programme_id
, ext
='mp4', entry_protocol
='m3u8_native',
252 m3u8_id
=supplier
, fatal
=False))
257 'format_id': supplier
or kind
or protocol
,
259 elif protocol
== 'rtmp':
260 application
= connection
.get('application', 'ondemand')
261 auth_string
= connection
.get('authString')
262 identifier
= connection
.get('identifier')
263 server
= connection
.get('server')
265 'url': '%s://%s/%s?%s' % (protocol
, server
, application
, auth_string
),
266 'play_path': identifier
,
267 'app': '%s?%s' % (application
, auth_string
),
268 'page_url': 'http://www.bbc.co.uk',
269 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
272 'format_id': supplier
,
276 def _extract_items(self
, playlist
):
277 return playlist
.findall('./{%s}item' % self
._EMP
_PLAYLIST
_NS
)
279 def _findall_ns(self
, element
, xpath
):
281 for ns
in self
._NAMESPACES
:
282 elements
.extend(element
.findall(xpath
% ns
))
285 def _extract_medias(self
, media_selection
):
286 error
= media_selection
.find('./{%s}error' % self
._MEDIASELECTION
_NS
)
288 media_selection
.find('./{%s}error' % self
._EMP
_PLAYLIST
_NS
)
289 if error
is not None:
290 raise BBCCoUkIE
.MediaSelectionError(error
.get('id'))
291 return self
._findall
_ns
(media_selection
, './{%s}media')
293 def _extract_connections(self
, media
):
294 return self
._findall
_ns
(media
, './{%s}connection')
296 def _extract_video(self
, media
, programme_id
):
298 vbr
= int_or_none(media
.get('bitrate'))
299 vcodec
= media
.get('encoding')
300 service
= media
.get('service')
301 width
= int_or_none(media
.get('width'))
302 height
= int_or_none(media
.get('height'))
303 file_size
= int_or_none(media
.get('media_file_size'))
304 for connection
in self
._extract
_connections
(media
):
305 conn_formats
= self
._extract
_connection
(connection
, programme_id
)
306 for format
in conn_formats
:
312 'filesize': file_size
,
315 format
['format_id'] = '%s_%s' % (service
, format
['format_id'])
316 formats
.extend(conn_formats
)
319 def _extract_audio(self
, media
, programme_id
):
321 abr
= int_or_none(media
.get('bitrate'))
322 acodec
= media
.get('encoding')
323 service
= media
.get('service')
324 for connection
in self
._extract
_connections
(media
):
325 conn_formats
= self
._extract
_connection
(connection
, programme_id
)
326 for format
in conn_formats
:
328 'format_id': '%s_%s' % (service
, format
['format_id']),
333 formats
.extend(conn_formats
)
336 def _get_subtitles(self
, media
, programme_id
):
338 for connection
in self
._extract
_connections
(media
):
339 captions
= self
._download
_xml
(connection
.get('href'), programme_id
, 'Downloading captions')
340 lang
= captions
.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
343 'url': connection
.get('href'),
349 def _raise_extractor_error(self
, media_selection_error
):
350 raise ExtractorError(
351 '%s returned error: %s' % (self
.IE_NAME
, media_selection_error
.id),
354 def _download_media_selector(self
, programme_id
):
355 last_exception
= None
356 for mediaselector_url
in self
._MEDIASELECTOR
_URLS
:
358 return self
._download
_media
_selector
_url
(
359 mediaselector_url
% programme_id
, programme_id
)
360 except BBCCoUkIE
.MediaSelectionError
as e
:
361 if e
.id in ('notukerror', 'geolocation', 'selectionunavailable'):
364 self
._raise
_extractor
_error
(e
)
365 self
._raise
_extractor
_error
(last_exception
)
367 def _download_media_selector_url(self
, url
, programme_id
=None):
369 media_selection
= self
._download
_xml
(
370 url
, programme_id
, 'Downloading media selection XML')
371 except ExtractorError
as ee
:
372 if isinstance(ee
.cause
, compat_HTTPError
) and ee
.cause
.code
in (403, 404):
373 media_selection
= compat_etree_fromstring(ee
.cause
.read().decode('utf-8'))
376 return self
._process
_media
_selector
(media_selection
, programme_id
)
378 def _process_media_selector(self
, media_selection
, programme_id
):
382 for media
in self
._extract
_medias
(media_selection
):
383 kind
= media
.get('kind')
385 formats
.extend(self
._extract
_audio
(media
, programme_id
))
386 elif kind
== 'video':
387 formats
.extend(self
._extract
_video
(media
, programme_id
))
388 elif kind
== 'captions':
389 subtitles
= self
.extract_subtitles(media
, programme_id
)
390 return formats
, subtitles
392 def _download_playlist(self
, playlist_id
):
394 playlist
= self
._download
_json
(
395 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id
,
396 playlist_id
, 'Downloading playlist JSON')
398 version
= playlist
.get('defaultAvailableVersion')
400 smp_config
= version
['smpConfig']
401 title
= smp_config
['title']
402 description
= smp_config
['summary']
403 for item
in smp_config
['items']:
405 if kind
!= 'programme' and kind
!= 'radioProgramme':
407 programme_id
= item
.get('vpid')
408 duration
= int_or_none(item
.get('duration'))
409 formats
, subtitles
= self
._download
_media
_selector
(programme_id
)
410 return programme_id
, title
, description
, duration
, formats
, subtitles
411 except ExtractorError
as ee
:
412 if not (isinstance(ee
.cause
, compat_HTTPError
) and ee
.cause
.code
== 404):
415 # fallback to legacy playlist
416 return self
._process
_legacy
_playlist
(playlist_id
)
418 def _process_legacy_playlist_url(self
, url
, display_id
):
419 playlist
= self
._download
_legacy
_playlist
_url
(url
, display_id
)
420 return self
._extract
_from
_legacy
_playlist
(playlist
, display_id
)
422 def _process_legacy_playlist(self
, playlist_id
):
423 return self
._process
_legacy
_playlist
_url
(
424 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id
, playlist_id
)
426 def _download_legacy_playlist_url(self
, url
, playlist_id
=None):
427 return self
._download
_xml
(
428 url
, playlist_id
, 'Downloading legacy playlist XML')
430 def _extract_from_legacy_playlist(self
, playlist
, playlist_id
):
431 no_items
= playlist
.find('./{%s}noItems' % self
._EMP
_PLAYLIST
_NS
)
432 if no_items
is not None:
433 reason
= no_items
.get('reason')
434 if reason
== 'preAvailability':
435 msg
= 'Episode %s is not yet available' % playlist_id
436 elif reason
== 'postAvailability':
437 msg
= 'Episode %s is no longer available' % playlist_id
438 elif reason
== 'noMedia':
439 msg
= 'Episode %s is not currently available' % playlist_id
441 msg
= 'Episode %s is not available: %s' % (playlist_id
, reason
)
442 raise ExtractorError(msg
, expected
=True)
444 for item
in self
._extract
_items
(playlist
):
445 kind
= item
.get('kind')
446 if kind
!= 'programme' and kind
!= 'radioProgramme':
448 title
= playlist
.find('./{%s}title' % self
._EMP
_PLAYLIST
_NS
).text
449 description_el
= playlist
.find('./{%s}summary' % self
._EMP
_PLAYLIST
_NS
)
450 description
= description_el
.text
if description_el
is not None else None
452 def get_programme_id(item
):
453 def get_from_attributes(item
):
454 for p
in('identifier', 'group'):
456 if value
and re
.match(r
'^[pb][\da-z]{7}$', value
):
458 get_from_attributes(item
)
459 mediator
= item
.find('./{%s}mediator' % self
._EMP
_PLAYLIST
_NS
)
460 if mediator
is not None:
461 return get_from_attributes(mediator
)
463 programme_id
= get_programme_id(item
)
464 duration
= int_or_none(item
.get('duration'))
467 formats
, subtitles
= self
._download
_media
_selector
(programme_id
)
469 formats
, subtitles
= self
._process
_media
_selector
(item
, playlist_id
)
470 programme_id
= playlist_id
472 return programme_id
, title
, description
, duration
, formats
, subtitles
474 def _real_extract(self
, url
):
475 group_id
= self
._match
_id
(url
)
477 webpage
= self
._download
_webpage
(url
, group_id
, 'Downloading video page')
482 tviplayer
= self
._search
_regex
(
483 r
'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
484 webpage
, 'player', default
=None)
487 player
= self
._parse
_json
(tviplayer
, group_id
).get('player', {})
488 duration
= int_or_none(player
.get('duration'))
489 programme_id
= player
.get('vpid')
492 programme_id
= self
._search
_regex
(
493 r
'"vpid"\s*:\s*"(%s)"' % self
._ID
_REGEX
, webpage
, 'vpid', fatal
=False, default
=None)
496 formats
, subtitles
= self
._download
_media
_selector
(programme_id
)
497 title
= self
._og
_search
_title
(webpage
, default
=None) or self
._html
_search
_regex
(
498 (r
'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>',
499 r
'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage
, 'title')
500 description
= self
._search
_regex
(
501 (r
'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
502 r
'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'),
503 webpage
, 'description', default
=None)
505 description
= self
._html
_search
_meta
('description', webpage
)
507 programme_id
, title
, description
, duration
, formats
, subtitles
= self
._download
_playlist
(group_id
)
509 self
._sort
_formats
(formats
)
514 'description': description
,
515 'thumbnail': self
._og
_search
_thumbnail
(webpage
, default
=None),
516 'duration': duration
,
518 'subtitles': subtitles
,
522 class BBCIE(BBCCoUkIE
):
525 _VALID_URL
= r
'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
527 _MEDIASELECTOR_URLS
= [
528 # Provides HQ HLS streams but fails with geolocation in some cases when it's
529 # even not geo restricted at all
530 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
531 # Provides more formats, namely direct mp4 links, but fails on some videos with
532 # notukerror for non UK (?) users (e.g.
533 # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
534 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
535 # Provides fewer formats, but works everywhere for everybody (hopefully)
536 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
540 # article with multiple videos embedded with data-playable containing vpids
541 'url': 'http://www.bbc.com/news/world-europe-32668511',
543 'id': 'world-europe-32668511',
544 'title': 'Russia stages massive WW2 parade despite Western boycott',
545 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
549 # article with multiple videos embedded with data-playable (more videos)
550 'url': 'http://www.bbc.com/news/business-28299555',
552 'id': 'business-28299555',
553 'title': 'Farnborough Airshow: Video highlights',
554 'description': 'BBC reports and video highlights at the Farnborough Airshow.',
559 # article with multiple videos embedded with `new SMP()`
561 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
563 'id': '3662a707-0af9-3149-963f-47bea720b460',
566 'playlist_count': 18,
568 # single video embedded with data-playable containing vpid
569 'url': 'http://www.bbc.com/news/world-europe-32041533',
573 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
574 'description': 'md5:2868290467291b37feda7863f7a83f54',
576 'timestamp': 1427219242,
577 'upload_date': '20150324',
581 'skip_download': True,
584 # article with single video embedded with data-playable containing XML playlist
585 # with direct video links as progressiveDownloadUrl (for now these are extracted)
586 # and playlist with f4m and m3u8 as streamingUrl
587 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
589 'id': '150615_telabyad_kentin_cogu',
591 'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
592 'timestamp': 1434397334,
593 'upload_date': '20150615',
596 'skip_download': True,
599 # single video embedded with data-playable containing XML playlists (regional section)
600 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
602 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
604 'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
605 'timestamp': 1434713142,
606 'upload_date': '20150619',
609 'skip_download': True,
612 # single video from video playlist embedded with vxp-playlist-data JSON
613 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
617 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
619 'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
622 'skip_download': True,
625 # single video story with digitalData
626 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
630 'title': 'Sri Lanka’s spicy secret',
631 'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
632 'timestamp': 1437674293,
633 'upload_date': '20150723',
637 'skip_download': True,
640 # single video story without digitalData
641 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
645 'title': 'Hyundai Santa Fe Sport: Rock star',
646 'description': 'md5:b042a26142c4154a6e472933cf20793d',
647 'timestamp': 1415867444,
648 'upload_date': '20141113',
652 'skip_download': True,
655 # single video with playlist.sxml URL in playlist param
656 'url': 'http://www.bbc.com/sport/0/football/33653409',
660 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
661 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
666 'skip_download': True,
669 # article with multiple videos embedded with playlist.sxml in playlist param
670 'url': 'http://www.bbc.com/sport/0/football/34475836',
673 'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
674 'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
678 # school report article with single video
679 'url': 'http://www.bbc.co.uk/schoolreport/35744779',
682 'title': 'School which breaks down barriers in Jerusalem',
686 # single video with playlist URL from weather section
687 'url': 'http://www.bbc.com/weather/features/33601775',
688 'only_matching': True,
690 # custom redirection to www.bbc.com
691 'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
692 'only_matching': True,
694 # single video article embedded with data-media-vpid
695 'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
696 'only_matching': True,
700 def suitable(cls
, url
):
701 return False if BBCCoUkIE
.suitable(url
) or BBCCoUkArticleIE
.suitable(url
) else super(BBCIE
, cls
).suitable(url
)
703 def _extract_from_media_meta(self
, media_meta
, video_id
):
704 # Direct links to media in media metadata (e.g.
705 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
706 # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
707 source_files
= media_meta
.get('sourceFiles')
711 'format_id': format_id
,
712 'ext': f
.get('encoding'),
713 'tbr': float_or_none(f
.get('bitrate'), 1000),
714 'filesize': int_or_none(f
.get('filesize')),
715 } for format_id
, f
in source_files
.items() if f
.get('url')], []
717 programme_id
= media_meta
.get('externalId')
719 return self
._download
_media
_selector
(programme_id
)
721 # Process playlist.sxml as legacy playlist
722 href
= media_meta
.get('href')
724 playlist
= self
._download
_legacy
_playlist
_url
(href
)
725 _
, _
, _
, _
, formats
, subtitles
= self
._extract
_from
_legacy
_playlist
(playlist
, video_id
)
726 return formats
, subtitles
730 def _extract_from_playlist_sxml(self
, url
, playlist_id
, timestamp
):
731 programme_id
, title
, description
, duration
, formats
, subtitles
= \
732 self
._process
_legacy
_playlist
_url
(url
, playlist_id
)
733 self
._sort
_formats
(formats
)
737 'description': description
,
738 'duration': duration
,
739 'timestamp': timestamp
,
741 'subtitles': subtitles
,
744 def _real_extract(self
, url
):
745 playlist_id
= self
._match
_id
(url
)
747 webpage
= self
._download
_webpage
(url
, playlist_id
)
749 json_ld_info
= self
._search
_json
_ld
(webpage
, playlist_id
, default
=None)
750 timestamp
= json_ld_info
.get('timestamp')
752 playlist_title
= json_ld_info
.get('title')
753 if not playlist_title
:
754 playlist_title
= self
._og
_search
_title
(
755 webpage
, default
=None) or self
._html
_search
_regex
(
756 r
'<title>(.+?)</title>', webpage
, 'playlist title', default
=None)
758 playlist_title
= re
.sub(r
'(.+)\s*-\s*BBC.*?$', r
'\1', playlist_title
).strip()
760 playlist_description
= json_ld_info
.get(
761 'description') or self
._og
_search
_description
(webpage
, default
=None)
764 timestamp
= parse_iso8601(self
._search
_regex
(
765 [r
'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
766 r
'itemprop="datePublished"[^>]+datetime="([^"]+)"',
767 r
'"datePublished":\s*"([^"]+)'],
768 webpage
, 'date', default
=None))
772 # article with multiple videos embedded with playlist.sxml (e.g.
773 # http://www.bbc.com/sport/0/football/34475836)
774 playlists
= re
.findall(r
'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage
)
775 playlists
.extend(re
.findall(r
'data-media-id="([^"]+/playlist\.sxml)"', webpage
))
778 self
._extract
_from
_playlist
_sxml
(playlist_url
, playlist_id
, timestamp
)
779 for playlist_url
in playlists
]
781 # news article with multiple videos embedded with data-playable
782 data_playables
= re
.findall(r
'data-playable=(["\'])({.+?}
)\
1', webpage)
784 for _, data_playable_json in data_playables:
785 data_playable = self._parse_json(
786 unescapeHTML(data_playable_json), playlist_id, fatal=False)
787 if not data_playable:
789 settings = data_playable.get('settings
', {})
791 # data-playable with video vpid in settings.playlistObject.items (e.g.
792 # http://www.bbc.com/news/world-us-canada-34473351)
793 playlist_object = settings.get('playlistObject
', {})
795 items = playlist_object.get('items
')
796 if items and isinstance(items, list):
797 title = playlist_object['title
']
798 description = playlist_object.get('summary
')
799 duration = int_or_none(items[0].get('duration
'))
800 programme_id = items[0].get('vpid
')
801 formats, subtitles = self._download_media_selector(programme_id)
802 self._sort_formats(formats)
806 'description
': description,
807 'timestamp
': timestamp,
808 'duration
': duration,
810 'subtitles
': subtitles,
813 # data-playable without vpid but with a playlist.sxml URLs
814 # in otherSettings.playlist (e.g.
815 # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
816 playlist = data_playable.get('otherSettings
', {}).get('playlist', {})
818 entries.append(self._extract_from_playlist_sxml(
819 playlist.get('progressiveDownloadUrl
'), playlist_id, timestamp))
822 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
824 # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
825 programme_id = self._search_regex(
826 [r'data
-(?
:video
-player|media
)-vpid
="(%s)"' % self._ID_REGEX,
827 r'<param
[^
>]+name
="externalIdentifier"[^
>]+value
="(%s)"' % self._ID_REGEX,
828 r'videoId\s
*:\s
*["\'](%s)["\']' % self._ID_REGEX],
829 webpage, 'vpid
', default=None)
832 formats, subtitles = self._download_media_selector(programme_id)
833 self._sort_formats(formats)
834 # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
835 digital_data = self._parse_json(
837 r'var\s
+digitalData\s
*=\s
*({.+?}
);?
\n', webpage, 'digital data
', default='{}'),
838 programme_id, fatal=False)
839 page_info = digital_data.get('page
', {}).get('pageInfo', {})
840 title = page_info.get('pageName
') or self._og_search_title(webpage)
841 description = page_info.get('description
') or self._og_search_description(webpage)
842 timestamp = parse_iso8601(page_info.get('publicationDate
')) or timestamp
846 'description
': description,
847 'timestamp
': timestamp,
849 'subtitles
': subtitles,
852 def extract_all(pattern):
853 return list(filter(None, map(
854 lambda s: self._parse_json(s, playlist_id, fatal=False),
855 re.findall(pattern, webpage))))
857 # Multiple video article (e.g.
858 # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
859 EMBED_URL = r'https?
://(?
:www\
.)?bbc\
.co\
.uk
/(?
:[^
/]+/)+%s(?
:\b[^
"]+)?' % self._ID_REGEX
861 for match in extract_all(r'new\s+SMP\(({.+?})\)'):
862 embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
863 if embed_url and re.match(EMBED_URL, embed_url):
864 entries.append(embed_url)
865 entries.extend(re.findall(
866 r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
868 return self.playlist_result(
869 [self.url_result(entry, 'BBCCoUk') for entry in entries],
870 playlist_id, playlist_title, playlist_description)
872 # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
873 medias = extract_all(r"data
-media
-meta
='({[^']+})'")
876 # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
877 media_asset = self._search_regex(
878 r'mediaAssetPage\.init\(\s*({.+?}), "/',
879 webpage, 'media asset
', default=None)
881 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
883 for video in media_asset_page.get('videos
', {}).values():
884 medias.extend(video.values())
887 # Multiple video playlist with single `now playing` entry (e.g.
888 # http://www.bbc.com/news/video_and_audio/must_see/33767813)
889 vxp_playlist = self._parse_json(
891 r'<script
[^
>]+class="vxp-playlist-data"[^
>]+type="application/json"[^
>]*>([^
<]+)</script
>',
892 webpage, 'playlist data
'),
895 for item in vxp_playlist:
896 media = item.get('media
')
899 playlist_medias.append(media)
900 # Download single video if found media with asset id matching the video id from URL
901 if item.get('advert
', {}).get('assetId
') == playlist_id:
904 # Fallback to the whole playlist
906 medias = playlist_medias
909 for num, media_meta in enumerate(medias, start=1):
910 formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
913 self._sort_formats(formats)
915 video_id = media_meta.get('externalId
')
917 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
919 title = media_meta.get('caption
')
921 title = playlist_title if len(medias) == 1 else '%s - Video
%s' % (playlist_title, num)
923 duration = int_or_none(media_meta.get('durationInSeconds
')) or parse_duration(media_meta.get('duration
'))
926 for image in media_meta.get('images
', {}).values():
927 images.extend(image.values())
928 if 'image
' in media_meta:
929 images.append(media_meta['image
'])
932 'url
': image.get('href
'),
933 'width
': int_or_none(image.get('width
')),
934 'height
': int_or_none(image.get('height
')),
935 } for image in images]
940 'thumbnails
': thumbnails,
941 'duration
': duration,
942 'timestamp
': timestamp,
944 'subtitles
': subtitles,
947 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
950 class BBCCoUkArticleIE(InfoExtractor):
951 _VALID_URL = r'https?
://www
.bbc
.co
.uk
/programmes
/articles
/(?P
<id>[a
-zA
-Z0
-9]+)'
952 IE_NAME = 'bbc
.co
.uk
:article
'
953 IE_DESC = 'BBC articles
'
956 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/articles
/3jNQLTMrPlYGTBn0WV6M2MS
/not-your
-typical
-role
-model
-ada
-lovelace
-the
-19th
-century
-programmer
',
958 'id': '3jNQLTMrPlYGTBn0WV6M2MS
',
959 'title
': 'Calculating Ada
: The Countess of Computing
- Not your typical role model
: Ada Lovelace the
19th century programmer
- BBC Four
',
960 'description
': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming
.',
963 'add_ie
': ['BBCCoUk
'],
966 def _real_extract(self, url):
967 playlist_id = self._match_id(url)
969 webpage = self._download_webpage(url, playlist_id)
971 title = self._og_search_title(webpage)
972 description = self._og_search_description(webpage).strip()
974 entries = [self.url_result(programme_url) for programme_url in re.findall(
975 r'<div
[^
>]+typeof
="Clip"[^
>]+resource
="([^"]+)"', webpage)]
977 return self.playlist_result(entries, playlist_id, title, description)