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
'[pb][\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,
238 _USP_RE
= r
'/([^/]+?)\.ism(?:\.hlsv2\.ism)?/[^/]+\.m3u8'
241 username
, password
= self
._get
_login
_info
()
245 login_page
= self
._download
_webpage
(
246 self
._LOGIN
_URL
, None, 'Downloading signin page')
248 login_form
= self
._hidden
_inputs
(login_page
)
251 'username': username
,
252 'password': password
,
255 post_url
= urljoin(self
._LOGIN
_URL
, self
._search
_regex
(
256 r
'<form[^>]+action=(["\'])(?P
<url
>.+?
)\
1', login_page,
257 'post url
', default=self._LOGIN_URL, group='url
'))
259 response, urlh = self._download_webpage_handle(
260 post_url, None, 'Logging
in', data=urlencode_postdata(login_form),
261 headers={'Referer': self._LOGIN_URL})
263 if self._LOGIN_URL in urlh.geturl():
264 error = clean_html(get_element_by_class('form
-message
', response))
266 raise ExtractorError(
267 'Unable to login
: %s' % error, expected=True)
268 raise ExtractorError('Unable to log
in')
270 def _real_initialize(self):
273 class MediaSelectionError(Exception):
274 def __init__(self, id):
277 def _extract_asx_playlist(self, connection, programme_id):
278 asx = self._download_xml(connection.get('href
'), programme_id, 'Downloading ASX playlist
')
279 return [ref.get('href
') for ref in asx.findall('./Entry
/ref
')]
281 def _extract_items(self, playlist):
282 return playlist.findall('./{%s}item
' % self._EMP_PLAYLIST_NS)
284 def _findall_ns(self, element, xpath):
286 for ns in self._NAMESPACES:
287 elements.extend(element.findall(xpath % ns))
290 def _extract_medias(self, media_selection):
291 error = media_selection.find('./{%s}error
' % self._MEDIASELECTION_NS)
293 media_selection.find('./{%s}error
' % self._EMP_PLAYLIST_NS)
294 if error is not None:
295 raise BBCCoUkIE.MediaSelectionError(error.get('id'))
296 return self._findall_ns(media_selection, './{%s}media
')
298 def _extract_connections(self, media):
299 return self._findall_ns(media, './{%s}connection
')
301 def _get_subtitles(self, media, programme_id):
303 for connection in self._extract_connections(media):
304 captions = self._download_xml(connection.get('href
'), programme_id, 'Downloading captions
')
305 lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang
', 'en
')
308 'url
': connection.get('href
'),
314 def _raise_extractor_error(self, media_selection_error):
315 raise ExtractorError(
316 '%s returned error
: %s' % (self.IE_NAME, media_selection_error.id),
319 def _download_media_selector(self, programme_id):
320 last_exception = None
321 for mediaselector_url in self._MEDIASELECTOR_URLS:
323 return self._download_media_selector_url(
324 mediaselector_url % programme_id, programme_id)
325 except BBCCoUkIE.MediaSelectionError as e:
326 if e.id in ('notukerror
', 'geolocation
', 'selectionunavailable
'):
329 self._raise_extractor_error(e)
330 self._raise_extractor_error(last_exception)
332 def _download_media_selector_url(self, url, programme_id=None):
334 media_selection = self._download_xml(
335 url, programme_id, 'Downloading media selection XML
')
336 except ExtractorError as ee:
337 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code in (403, 404):
338 media_selection = compat_etree_fromstring(ee.cause.read().decode('utf
-8'))
341 return self._process_media_selector(media_selection, programme_id)
343 def _process_media_selector(self, media_selection, programme_id):
348 for media in self._extract_medias(media_selection):
349 kind = media.get('kind
')
350 if kind in ('video
', 'audio
'):
351 bitrate = int_or_none(media.get('bitrate
'))
352 encoding = media.get('encoding
')
353 service = media.get('service
')
354 width = int_or_none(media.get('width
'))
355 height = int_or_none(media.get('height
'))
356 file_size = int_or_none(media.get('media_file_size
'))
357 for connection in self._extract_connections(media):
358 href = connection.get('href
')
363 conn_kind = connection.get('kind
')
364 protocol = connection.get('protocol
')
365 supplier = connection.get('supplier
')
366 transfer_format = connection.get('transferFormat
')
367 format_id = supplier or conn_kind or protocol
369 format_id = '%s_%s' % (service, format_id)
371 if supplier == 'asx
':
372 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
375 'format_id
': 'ref
%s_%s' % (i, format_id),
377 elif transfer_format == 'dash
':
378 formats.extend(self._extract_mpd_formats(
379 href, programme_id, mpd_id=format_id, fatal=False))
380 elif transfer_format == 'hls
':
381 formats.extend(self._extract_m3u8_formats(
382 href, programme_id, ext='mp4
', entry_protocol='m3u8_native
',
383 m3u8_id=format_id, fatal=False))
384 if re.search(self._USP_RE, href):
385 usp_formats = self._extract_m3u8_formats(
386 re.sub(self._USP_RE, r'/\
1.ism
/\
1.m3u8
', href),
387 programme_id, ext='mp4
', entry_protocol='m3u8_native
',
388 m3u8_id=format_id, fatal=False)
389 for f in usp_formats:
390 if f.get('height
') and f['height
'] > 720:
393 elif transfer_format == 'hds
':
394 formats.extend(self._extract_f4m_formats(
395 href, programme_id, f4m_id=format_id, fatal=False))
397 if not service and not supplier and bitrate:
398 format_id += '-%d' % bitrate
400 'format_id
': format_id,
401 'filesize
': file_size,
416 if protocol in ('http
', 'https
'):
421 elif protocol == 'rtmp
':
422 application = connection.get('application
', 'ondemand
')
423 auth_string = connection.get('authString
')
424 identifier = connection.get('identifier
')
425 server = connection.get('server
')
427 'url
': '%s://%s/%s?
%s' % (protocol, server, application, auth_string),
428 'play_path
': identifier,
429 'app
': '%s?
%s' % (application, auth_string),
430 'page_url
': 'http
://www
.bbc
.co
.uk
',
431 'player_url
': 'http
://www
.bbc
.co
.uk
/emp
/releases
/iplayer
/revisions
/617463_618125_4/617463_618125_4_emp
.swf
',
438 elif kind == 'captions
':
439 subtitles = self.extract_subtitles(media, programme_id)
440 return formats, subtitles
442 def _download_playlist(self, playlist_id):
444 playlist = self._download_json(
445 'http
://www
.bbc
.co
.uk
/programmes
/%s/playlist
.json
' % playlist_id,
446 playlist_id, 'Downloading playlist JSON
')
448 version = playlist.get('defaultAvailableVersion
')
450 smp_config = version['smpConfig
']
451 title = smp_config['title
']
452 description = smp_config['summary
']
453 for item in smp_config['items
']:
455 if kind not in ('programme
', 'radioProgramme
'):
457 programme_id = item.get('vpid
')
458 duration = int_or_none(item.get('duration
'))
459 formats, subtitles = self._download_media_selector(programme_id)
460 return programme_id, title, description, duration, formats, subtitles
461 except ExtractorError as ee:
462 if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
465 # fallback to legacy playlist
466 return self._process_legacy_playlist(playlist_id)
468 def _process_legacy_playlist_url(self, url, display_id):
469 playlist = self._download_legacy_playlist_url(url, display_id)
470 return self._extract_from_legacy_playlist(playlist, display_id)
472 def _process_legacy_playlist(self, playlist_id):
473 return self._process_legacy_playlist_url(
474 'http
://www
.bbc
.co
.uk
/iplayer
/playlist
/%s' % playlist_id, playlist_id)
476 def _download_legacy_playlist_url(self, url, playlist_id=None):
477 return self._download_xml(
478 url, playlist_id, 'Downloading legacy playlist XML
')
480 def _extract_from_legacy_playlist(self, playlist, playlist_id):
481 no_items = playlist.find('./{%s}noItems
' % self._EMP_PLAYLIST_NS)
482 if no_items is not None:
483 reason = no_items.get('reason
')
484 if reason == 'preAvailability
':
485 msg = 'Episode
%s is not yet available
' % playlist_id
486 elif reason == 'postAvailability
':
487 msg = 'Episode
%s is no longer available
' % playlist_id
488 elif reason == 'noMedia
':
489 msg = 'Episode
%s is not currently available
' % playlist_id
491 msg = 'Episode
%s is not available
: %s' % (playlist_id, reason)
492 raise ExtractorError(msg, expected=True)
494 for item in self._extract_items(playlist):
495 kind = item.get('kind
')
496 if kind not in ('programme
', 'radioProgramme
'):
498 title = playlist.find('./{%s}title
' % self._EMP_PLAYLIST_NS).text
499 description_el = playlist.find('./{%s}summary
' % self._EMP_PLAYLIST_NS)
500 description = description_el.text if description_el is not None else None
502 def get_programme_id(item):
503 def get_from_attributes(item):
504 for p in('identifier
', 'group
'):
506 if value and re.match(r'^
[pb
][\da
-z
]{7}$
', value):
508 get_from_attributes(item)
509 mediator = item.find('./{%s}mediator
' % self._EMP_PLAYLIST_NS)
510 if mediator is not None:
511 return get_from_attributes(mediator)
513 programme_id = get_programme_id(item)
514 duration = int_or_none(item.get('duration
'))
517 formats, subtitles = self._download_media_selector(programme_id)
519 formats, subtitles = self._process_media_selector(item, playlist_id)
520 programme_id = playlist_id
522 return programme_id, title, description, duration, formats, subtitles
524 def _real_extract(self, url):
525 group_id = self._match_id(url)
527 webpage = self._download_webpage(url, group_id, 'Downloading video page
')
529 error = self._search_regex(
530 r'<div
\b[^
>]+\bclass
=["\']smp__message delta["\'][^
>]*>([^
<]+)<',
531 webpage, 'error
', default=None)
533 raise ExtractorError(error, expected=True)
538 tviplayer = self._search_regex(
539 r'mediator\
.bind\
(({.+?}
)\s
*,\s
*document\
.getElementById
',
540 webpage, 'player
', default=None)
543 player = self._parse_json(tviplayer, group_id).get('player
', {})
544 duration = int_or_none(player.get('duration
'))
545 programme_id = player.get('vpid
')
548 programme_id = self._search_regex(
549 r'"vpid"\s
*:\s
*"(%s)"' % self._ID_REGEX, webpage, 'vpid
', fatal=False, default=None)
552 formats, subtitles = self._download_media_selector(programme_id)
553 title = self._og_search_title(webpage, default=None) or self._html_search_regex(
554 (r'<h2
[^
>]+id="parent-title"[^
>]*>(.+?
)</h2
>',
555 r'<div
[^
>]+class="info"[^
>]*>\s
*<h1
>(.+?
)</h1
>'), webpage, 'title
')
556 description = self._search_regex(
557 (r'<p
class="[^"]*medium
-description
[^
"]*">([^
<]+)</p
>',
558 r'<div
[^
>]+class="info_+synopsis"[^
>]*>([^
<]+)</div
>'),
559 webpage, 'description
', default=None)
561 description = self._html_search_meta('description
', webpage)
563 programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
565 self._sort_formats(formats)
570 'description
': description,
571 'thumbnail
': self._og_search_thumbnail(webpage, default=None),
572 'duration
': duration,
574 'subtitles
': subtitles,
578 class BBCIE(BBCCoUkIE):
581 _VALID_URL = r'https?
://(?
:www\
.)?bbc\
.(?
:com|co\
.uk
)/(?
:[^
/]+/)+(?P
<id>[^
/#?]+)'
583 _MEDIASELECTOR_URLS
= [
584 # Provides HQ HLS streams but fails with geolocation in some cases when it's
585 # even not geo restricted at all
586 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
587 # Provides more formats, namely direct mp4 links, but fails on some videos with
588 # notukerror for non UK (?) users (e.g.
589 # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
590 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
591 # Provides fewer formats, but works everywhere for everybody (hopefully)
592 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
596 # article with multiple videos embedded with data-playable containing vpids
597 'url': 'http://www.bbc.com/news/world-europe-32668511',
599 'id': 'world-europe-32668511',
600 'title': 'Russia stages massive WW2 parade despite Western boycott',
601 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
605 # article with multiple videos embedded with data-playable (more videos)
606 'url': 'http://www.bbc.com/news/business-28299555',
608 'id': 'business-28299555',
609 'title': 'Farnborough Airshow: Video highlights',
610 'description': 'BBC reports and video highlights at the Farnborough Airshow.',
615 # article with multiple videos embedded with `new SMP()`
617 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
619 'id': '3662a707-0af9-3149-963f-47bea720b460',
622 'playlist_count': 18,
624 # single video embedded with data-playable containing vpid
625 'url': 'http://www.bbc.com/news/world-europe-32041533',
629 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
630 'description': 'md5:2868290467291b37feda7863f7a83f54',
632 'timestamp': 1427219242,
633 'upload_date': '20150324',
637 'skip_download': True,
640 # article with single video embedded with data-playable containing XML playlist
641 # with direct video links as progressiveDownloadUrl (for now these are extracted)
642 # and playlist with f4m and m3u8 as streamingUrl
643 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
645 'id': '150615_telabyad_kentin_cogu',
647 'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
648 'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
649 'timestamp': 1434397334,
650 'upload_date': '20150615',
653 'skip_download': True,
656 # single video embedded with data-playable containing XML playlists (regional section)
657 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
659 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
661 'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
662 'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
663 'timestamp': 1434713142,
664 'upload_date': '20150619',
667 'skip_download': True,
670 # single video from video playlist embedded with vxp-playlist-data JSON
671 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
675 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
677 'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
680 'skip_download': True,
683 # single video story with digitalData
684 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
688 'title': 'Sri Lanka’s spicy secret',
689 'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
690 'timestamp': 1437674293,
691 'upload_date': '20150723',
695 'skip_download': True,
698 # single video story without digitalData
699 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
703 'title': 'Hyundai Santa Fe Sport: Rock star',
704 'description': 'md5:b042a26142c4154a6e472933cf20793d',
705 'timestamp': 1415867444,
706 'upload_date': '20141113',
710 'skip_download': True,
713 # single video embedded with Morph
714 'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
718 'title': "Nigeria v Japan - Men's First Round",
719 'description': 'Live coverage of the first round from Group B at the Amazonia Arena.',
721 'uploader': 'BBC Sport',
722 'uploader_id': 'bbc_sport',
726 'skip_download': True,
728 'skip': 'Georestricted to UK',
730 # single video with playlist.sxml URL in playlist param
731 'url': 'http://www.bbc.com/sport/0/football/33653409',
735 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
736 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
741 'skip_download': True,
744 # article with multiple videos embedded with playlist.sxml in playlist param
745 'url': 'http://www.bbc.com/sport/0/football/34475836',
748 'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
749 'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
753 # school report article with single video
754 'url': 'http://www.bbc.co.uk/schoolreport/35744779',
757 'title': 'School which breaks down barriers in Jerusalem',
761 # single video with playlist URL from weather section
762 'url': 'http://www.bbc.com/weather/features/33601775',
763 'only_matching': True,
765 # custom redirection to www.bbc.com
766 'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
767 'only_matching': True,
769 # single video article embedded with data-media-vpid
770 'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
771 'only_matching': True,
775 def suitable(cls
, url
):
776 EXCLUDE_IE
= (BBCCoUkIE
, BBCCoUkArticleIE
, BBCCoUkIPlayerPlaylistIE
, BBCCoUkPlaylistIE
)
777 return (False if any(ie
.suitable(url
) for ie
in EXCLUDE_IE
)
778 else super(BBCIE
, cls
).suitable(url
))
780 def _extract_from_media_meta(self
, media_meta
, video_id
):
781 # Direct links to media in media metadata (e.g.
782 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
783 # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
784 source_files
= media_meta
.get('sourceFiles')
788 'format_id': format_id
,
789 'ext': f
.get('encoding'),
790 'tbr': float_or_none(f
.get('bitrate'), 1000),
791 'filesize': int_or_none(f
.get('filesize')),
792 } for format_id
, f
in source_files
.items() if f
.get('url')], []
794 programme_id
= media_meta
.get('externalId')
796 return self
._download
_media
_selector
(programme_id
)
798 # Process playlist.sxml as legacy playlist
799 href
= media_meta
.get('href')
801 playlist
= self
._download
_legacy
_playlist
_url
(href
)
802 _
, _
, _
, _
, formats
, subtitles
= self
._extract
_from
_legacy
_playlist
(playlist
, video_id
)
803 return formats
, subtitles
807 def _extract_from_playlist_sxml(self
, url
, playlist_id
, timestamp
):
808 programme_id
, title
, description
, duration
, formats
, subtitles
= \
809 self
._process
_legacy
_playlist
_url
(url
, playlist_id
)
810 self
._sort
_formats
(formats
)
814 'description': description
,
815 'duration': duration
,
816 'timestamp': timestamp
,
818 'subtitles': subtitles
,
821 def _real_extract(self
, url
):
822 playlist_id
= self
._match
_id
(url
)
824 webpage
= self
._download
_webpage
(url
, playlist_id
)
826 json_ld_info
= self
._search
_json
_ld
(webpage
, playlist_id
, default
={})
827 timestamp
= json_ld_info
.get('timestamp')
829 playlist_title
= json_ld_info
.get('title')
830 if not playlist_title
:
831 playlist_title
= self
._og
_search
_title
(
832 webpage
, default
=None) or self
._html
_search
_regex
(
833 r
'<title>(.+?)</title>', webpage
, 'playlist title', default
=None)
835 playlist_title
= re
.sub(r
'(.+)\s*-\s*BBC.*?$', r
'\1', playlist_title
).strip()
837 playlist_description
= json_ld_info
.get(
838 'description') or self
._og
_search
_description
(webpage
, default
=None)
841 timestamp
= parse_iso8601(self
._search
_regex
(
842 [r
'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
843 r
'itemprop="datePublished"[^>]+datetime="([^"]+)"',
844 r
'"datePublished":\s*"([^"]+)'],
845 webpage
, 'date', default
=None))
849 # article with multiple videos embedded with playlist.sxml (e.g.
850 # http://www.bbc.com/sport/0/football/34475836)
851 playlists
= re
.findall(r
'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage
)
852 playlists
.extend(re
.findall(r
'data-media-id="([^"]+/playlist\.sxml)"', webpage
))
855 self
._extract
_from
_playlist
_sxml
(playlist_url
, playlist_id
, timestamp
)
856 for playlist_url
in playlists
]
858 # news article with multiple videos embedded with data-playable
859 data_playables
= re
.findall(r
'data-playable=(["\'])({.+?}
)\
1', webpage)
861 for _, data_playable_json in data_playables:
862 data_playable = self._parse_json(
863 unescapeHTML(data_playable_json), playlist_id, fatal=False)
864 if not data_playable:
866 settings = data_playable.get('settings
', {})
868 # data-playable with video vpid in settings.playlistObject.items (e.g.
869 # http://www.bbc.com/news/world-us-canada-34473351)
870 playlist_object = settings.get('playlistObject
', {})
872 items = playlist_object.get('items
')
873 if items and isinstance(items, list):
874 title = playlist_object['title
']
875 description = playlist_object.get('summary
')
876 duration = int_or_none(items[0].get('duration
'))
877 programme_id = items[0].get('vpid
')
878 formats, subtitles = self._download_media_selector(programme_id)
879 self._sort_formats(formats)
883 'description
': description,
884 'timestamp
': timestamp,
885 'duration
': duration,
887 'subtitles
': subtitles,
890 # data-playable without vpid but with a playlist.sxml URLs
891 # in otherSettings.playlist (e.g.
892 # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
893 playlist = data_playable.get('otherSettings
', {}).get('playlist', {})
896 for key in ('streaming
', 'progressiveDownload
'):
897 playlist_url = playlist.get('%sUrl
' % key)
901 info = self._extract_from_playlist_sxml(
902 playlist_url, playlist_id, timestamp)
906 entry['title
'] = info['title
']
907 entry['formats
'].extend(info['formats
'])
908 except Exception as e:
909 # Some playlist URL may fail with 500, at the same time
910 # the other one may work fine (e.g.
911 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
912 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
916 self._sort_formats(entry['formats
'])
917 entries.append(entry)
920 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
922 # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
923 programme_id = self._search_regex(
924 [r'data
-(?
:video
-player|media
)-vpid
="(%s)"' % self._ID_REGEX,
925 r'<param
[^
>]+name
="externalIdentifier"[^
>]+value
="(%s)"' % self._ID_REGEX,
926 r'videoId\s
*:\s
*["\'](%s)["\']' % self._ID_REGEX],
927 webpage, 'vpid
', default=None)
930 formats, subtitles = self._download_media_selector(programme_id)
931 self._sort_formats(formats)
932 # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
933 digital_data = self._parse_json(
935 r'var\s
+digitalData\s
*=\s
*({.+?}
);?
\n', webpage, 'digital data
', default='{}'),
936 programme_id, fatal=False)
937 page_info = digital_data.get('page
', {}).get('pageInfo', {})
938 title = page_info.get('pageName
') or self._og_search_title(webpage)
939 description = page_info.get('description
') or self._og_search_description(webpage)
940 timestamp = parse_iso8601(page_info.get('publicationDate
')) or timestamp
944 'description
': description,
945 'timestamp
': timestamp,
947 'subtitles
': subtitles,
950 # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
951 # There are several setPayload calls may be present but the video
952 # seems to be always related to the first one
953 morph_payload = self._parse_json(
955 r'Morph\
.setPayload\
([^
,]+,\s
*({.+?}
)\
);',
956 webpage, 'morph payload
', default='{}'),
957 playlist_id, fatal=False)
959 components = try_get(morph_payload, lambda x: x['body
']['components
'], list) or []
960 for component in components:
961 if not isinstance(component, dict):
963 lead_media = try_get(component, lambda x: x['props
']['leadMedia
'], dict)
966 identifiers = lead_media.get('identifiers
')
967 if not identifiers or not isinstance(identifiers, dict):
969 programme_id = identifiers.get('vpid
') or identifiers.get('playablePid
')
972 title = lead_media.get('title
') or self._og_search_title(webpage)
973 formats, subtitles = self._download_media_selector(programme_id)
974 self._sort_formats(formats)
975 description = lead_media.get('summary
')
976 uploader = lead_media.get('masterBrand
')
977 uploader_id = lead_media.get('mid
')
979 duration_d = lead_media.get('duration
')
980 if isinstance(duration_d, dict):
981 duration = parse_duration(dict_get(
982 duration_d, ('rawDuration
', 'formattedDuration
', 'spokenDuration
')))
986 'description
': description,
987 'duration
': duration,
988 'uploader
': uploader,
989 'uploader_id
': uploader_id,
991 'subtitles
': subtitles,
994 def extract_all(pattern):
995 return list(filter(None, map(
996 lambda s: self._parse_json(s, playlist_id, fatal=False),
997 re.findall(pattern, webpage))))
999 # Multiple video article (e.g.
1000 # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
1001 EMBED_URL = r'https?
://(?
:www\
.)?bbc\
.co\
.uk
/(?
:[^
/]+/)+%s(?
:\b[^
"]+)?' % self._ID_REGEX
1003 for match in extract_all(r'new\s+SMP\(({.+?})\)'):
1004 embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
1005 if embed_url and re.match(EMBED_URL, embed_url):
1006 entries.append(embed_url)
1007 entries.extend(re.findall(
1008 r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
1010 return self.playlist_result(
1011 [self.url_result(entry_, 'BBCCoUk') for entry_ in entries],
1012 playlist_id, playlist_title, playlist_description)
1014 # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
1015 medias = extract_all(r"data
-media
-meta
='({[^']+})'")
1018 # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
1019 media_asset = self._search_regex(
1020 r'mediaAssetPage\.init\(\s*({.+?}), "/',
1021 webpage, 'media asset
', default=None)
1023 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
1025 for video in media_asset_page.get('videos
', {}).values():
1026 medias.extend(video.values())
1029 # Multiple video playlist with single `now playing` entry (e.g.
1030 # http://www.bbc.com/news/video_and_audio/must_see/33767813)
1031 vxp_playlist = self._parse_json(
1033 r'<script
[^
>]+class="vxp-playlist-data"[^
>]+type="application/json"[^
>]*>([^
<]+)</script
>',
1034 webpage, 'playlist data
'),
1036 playlist_medias = []
1037 for item in vxp_playlist:
1038 media = item.get('media
')
1041 playlist_medias.append(media)
1042 # Download single video if found media with asset id matching the video id from URL
1043 if item.get('advert
', {}).get('assetId
') == playlist_id:
1046 # Fallback to the whole playlist
1048 medias = playlist_medias
1051 for num, media_meta in enumerate(medias, start=1):
1052 formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
1055 self._sort_formats(formats)
1057 video_id = media_meta.get('externalId
')
1059 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
1061 title = media_meta.get('caption
')
1063 title = playlist_title if len(medias) == 1 else '%s - Video
%s' % (playlist_title, num)
1065 duration = int_or_none(media_meta.get('durationInSeconds
')) or parse_duration(media_meta.get('duration
'))
1068 for image in media_meta.get('images
', {}).values():
1069 images.extend(image.values())
1070 if 'image
' in media_meta:
1071 images.append(media_meta['image
'])
1074 'url
': image.get('href
'),
1075 'width
': int_or_none(image.get('width
')),
1076 'height
': int_or_none(image.get('height
')),
1077 } for image in images]
1082 'thumbnails
': thumbnails,
1083 'duration
': duration,
1084 'timestamp
': timestamp,
1086 'subtitles
': subtitles,
1089 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
1092 class BBCCoUkArticleIE(InfoExtractor):
1093 _VALID_URL = r'https?
://(?
:www\
.)?bbc\
.co\
.uk
/programmes
/articles
/(?P
<id>[a
-zA
-Z0
-9]+)'
1094 IE_NAME = 'bbc
.co
.uk
:article
'
1095 IE_DESC = 'BBC articles
'
1098 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/articles
/3jNQLTMrPlYGTBn0WV6M2MS
/not-your
-typical
-role
-model
-ada
-lovelace
-the
-19th
-century
-programmer
',
1100 'id': '3jNQLTMrPlYGTBn0WV6M2MS
',
1101 'title
': 'Calculating Ada
: The Countess of Computing
- Not your typical role model
: Ada Lovelace the
19th century programmer
- BBC Four
',
1102 'description
': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming
.',
1104 'playlist_count
': 4,
1105 'add_ie
': ['BBCCoUk
'],
1108 def _real_extract(self, url):
1109 playlist_id = self._match_id(url)
1111 webpage = self._download_webpage(url, playlist_id)
1113 title = self._og_search_title(webpage)
1114 description = self._og_search_description(webpage).strip()
1116 entries = [self.url_result(programme_url) for programme_url in re.findall(
1117 r'<div
[^
>]+typeof
="Clip"[^
>]+resource
="([^"]+)"', webpage)]
1119 return self.playlist_result(entries, playlist_id, title, description)
1122 class BBCCoUkPlaylistBaseIE(InfoExtractor):
1123 def _entries(self, webpage, url, playlist_id):
1124 single_page = 'page' in compat_urlparse.parse_qs(
1125 compat_urlparse.urlparse(url).query)
1126 for page_num in itertools.count(2):
1127 for video_id in re.findall(
1128 self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage):
1129 yield self.url_result(
1130 self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
1133 next_page = self._search_regex(
1134 r'<li[^>]+class=(["\'])pagination_
+next\
1[^
>]*><a
[^
>]+href
=(["\'])(?P<url>(?:(?!\2).)+)\2',
1135 webpage, 'next page url', default=None, group='url')
1138 webpage = self._download_webpage(
1139 compat_urlparse.urljoin(url, next_page), playlist_id,
1140 'Downloading page %d' % page_num, page_num)
1142 def _real_extract(self, url):
1143 playlist_id = self._match_id(url)
1145 webpage = self._download_webpage(url, playlist_id)
1147 title, description = self._extract_title_and_description(webpage)
1149 return self.playlist_result(
1150 self._entries(webpage, url, playlist_id),
1151 playlist_id, title, description)
1154 class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
1155 IE_NAME = 'bbc.co.uk:iplayer:playlist'
1156 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
1157 _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
1158 _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
1160 'url
': 'http
://www
.bbc
.co
.uk
/iplayer
/episodes
/b05rcz9v
',
1163 'title
': 'The Disappearance
',
1164 'description
': 'French thriller serial about a missing teenager
.',
1166 'playlist_mincount
': 6,
1167 'skip
': 'This programme
is not currently available on BBC iPlayer
',
1169 # Available for over a year unlike 30 days for most other programmes
1170 'url
': 'http
://www
.bbc
.co
.uk
/iplayer
/group
/p02tcc32
',
1173 'title
': 'Bohemian Icons
',
1174 'description
': 'md5
:683e901041b2fe9ba596f2ab04c4dbe7
',
1176 'playlist_mincount
': 10,
1179 def _extract_title_and_description(self, webpage):
1180 title = self._search_regex(r'<h1
>([^
<]+)</h1
>', webpage, 'title
', fatal=False)
1181 description = self._search_regex(
1182 r'<p
[^
>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
1183 webpage, 'description', fatal=False, group='value')
1184 return title, description
1187 class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
1188 IE_NAME = 'bbc.co.uk:playlist'
1189 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
1190 _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
1191 _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
1193 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b05rcz9v
/clips
',
1196 'title
': 'The Disappearance
- Clips
- BBC Four
',
1197 'description
': 'French thriller serial about a missing teenager
.',
1199 'playlist_mincount
': 7,
1201 # multipage playlist, explicit page
1202 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b00mfl7n
/clips?page
=1',
1205 'title
': 'Frozen Planet
- Clips
- BBC One
',
1206 'description
': 'md5
:65dcbf591ae628dafe32aa6c4a4a0d8c
',
1208 'playlist_mincount
': 24,
1210 # multipage playlist, all pages
1211 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b00mfl7n
/clips
',
1214 'title
': 'Frozen Planet
- Clips
- BBC One
',
1215 'description
': 'md5
:65dcbf591ae628dafe32aa6c4a4a0d8c
',
1217 'playlist_mincount
': 142,
1219 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b05rcz9v
/broadcasts
/2016/06',
1220 'only_matching
': True,
1222 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b05rcz9v
/clips
',
1223 'only_matching
': True,
1225 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b055jkys
/episodes
/player
',
1226 'only_matching
': True,
1229 def _extract_title_and_description(self, webpage):
1230 title = self._og_search_title(webpage, fatal=False)
1231 description = self._og_search_description(webpage)
1232 return title, description