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)[/#]|
42 (?P<id>%s)(?!/(?:episodes|broadcasts|clips))
45 _LOGIN_URL
= 'https://account.bbc.com/signin'
46 _NETRC_MACHINE
= 'bbc'
48 _MEDIASELECTOR_URLS
= [
49 # Provides HQ HLS streams with even better quality that pc mediaset but fails
50 # with geolocation in some cases when it's even not geo restricted at all (e.g.
51 # http://www.bbc.co.uk/programmes/b06bp7lf). Also may fail with selectionunavailable.
52 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
53 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
56 _MEDIASELECTION_NS
= 'http://bbc.co.uk/2008/mp/mediaselection'
57 _EMP_PLAYLIST_NS
= 'http://bbc.co.uk/2008/emp/playlist'
66 'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
70 'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
71 'description': 'The Canadian poet and songwriter reflects on his musical career.',
75 'skip_download': True,
79 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
83 'title': 'The Man in Black: Series 3: The Printed Name',
84 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
89 'skip_download': True,
91 'skip': 'Episode is no longer available on BBC iPlayer Radio',
94 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
98 'title': 'The Voice UK: Series 3: Blind Auditions 5',
99 '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.',
104 'skip_download': True,
106 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
109 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
113 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
114 'description': '2. Invasion',
119 'skip_download': True,
121 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
123 'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
127 'title': 'Pete Tong, The Essential New Tune Special',
128 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
133 'skip_download': True,
135 'skip': 'Episode is no longer available on BBC iPlayer Radio',
137 'url': 'http://www.bbc.co.uk/music/clips/p022h44b',
142 'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
143 'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
148 'skip_download': True,
151 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
156 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
157 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
162 'skip_download': True,
165 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
169 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
170 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
175 'skip_download': True,
177 'skip': 'geolocation',
179 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
183 '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.',
184 'title': 'Royal Academy Summer Exhibition',
189 'skip_download': True,
191 'skip': 'geolocation',
193 # iptv-all mediaset fails with geolocation however there is no geo restriction
194 # for this programme at all
195 'url': 'http://www.bbc.co.uk/programmes/b06rkn85',
199 'title': "Best of the Mini-Mixes 2015: Part 3, Annie Mac's Friday Night - BBC Radio 1",
200 'description': "Annie has part three in the Best of the Mini-Mixes 2015, plus the year's Most Played!",
204 'skip_download': True,
206 'skip': 'Now it\'s really geo-restricted',
208 # compact player (https://github.com/rg3/youtube-dl/issues/8147)
209 'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
213 'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
214 'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
218 'skip_download': True,
221 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
222 'only_matching': True,
224 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
225 'only_matching': True,
227 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
228 'only_matching': True,
230 'url': 'http://www.bbc.co.uk/radio/player/p03cchwf',
231 'only_matching': True,
233 'url': 'https://www.bbc.co.uk/music/audiovideo/popular#p055bc55',
234 'only_matching': True,
237 _USP_RE
= r
'/([^/]+?)\.ism(?:\.hlsv2\.ism)?/[^/]+\.m3u8'
240 username
, password
= self
._get
_login
_info
()
244 login_page
= self
._download
_webpage
(
245 self
._LOGIN
_URL
, None, 'Downloading signin page')
247 login_form
= self
._hidden
_inputs
(login_page
)
250 'username': username
,
251 'password': password
,
254 post_url
= urljoin(self
._LOGIN
_URL
, self
._search
_regex
(
255 r
'<form[^>]+action=(["\'])(?P
<url
>.+?
)\
1', login_page,
256 'post url
', default=self._LOGIN_URL, group='url
'))
258 response, urlh = self._download_webpage_handle(
259 post_url, None, 'Logging
in', data=urlencode_postdata(login_form),
260 headers={'Referer': self._LOGIN_URL})
262 if self._LOGIN_URL in urlh.geturl():
263 error = clean_html(get_element_by_class('form
-message
', response))
265 raise ExtractorError(
266 'Unable to login
: %s' % error, expected=True)
267 raise ExtractorError('Unable to log
in')
269 def _real_initialize(self):
272 class MediaSelectionError(Exception):
273 def __init__(self, id):
276 def _extract_asx_playlist(self, connection, programme_id):
277 asx = self._download_xml(connection.get('href
'), programme_id, 'Downloading ASX playlist
')
278 return [ref.get('href
') for ref in asx.findall('./Entry
/ref
')]
280 def _extract_items(self, playlist):
281 return playlist.findall('./{%s}item
' % self._EMP_PLAYLIST_NS)
283 def _findall_ns(self, element, xpath):
285 for ns in self._NAMESPACES:
286 elements.extend(element.findall(xpath % ns))
289 def _extract_medias(self, media_selection):
290 error = media_selection.find('./{%s}error
' % self._MEDIASELECTION_NS)
292 media_selection.find('./{%s}error
' % self._EMP_PLAYLIST_NS)
293 if error is not None:
294 raise BBCCoUkIE.MediaSelectionError(error.get('id'))
295 return self._findall_ns(media_selection, './{%s}media
')
297 def _extract_connections(self, media):
298 return self._findall_ns(media, './{%s}connection
')
300 def _get_subtitles(self, media, programme_id):
302 for connection in self._extract_connections(media):
303 captions = self._download_xml(connection.get('href
'), programme_id, 'Downloading captions
')
304 lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang
', 'en
')
307 'url
': connection.get('href
'),
313 def _raise_extractor_error(self, media_selection_error):
314 raise ExtractorError(
315 '%s returned error
: %s' % (self.IE_NAME, media_selection_error.id),
318 def _download_media_selector(self, programme_id):
319 last_exception = None
320 for mediaselector_url in self._MEDIASELECTOR_URLS:
322 return self._download_media_selector_url(
323 mediaselector_url % programme_id, programme_id)
324 except BBCCoUkIE.MediaSelectionError as e:
325 if e.id in ('notukerror
', 'geolocation
', 'selectionunavailable
'):
328 self._raise_extractor_error(e)
329 self._raise_extractor_error(last_exception)
331 def _download_media_selector_url(self, url, programme_id=None):
333 media_selection = self._download_xml(
334 url, programme_id, 'Downloading media selection XML
')
335 except ExtractorError as ee:
336 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code in (403, 404):
337 media_selection = compat_etree_fromstring(ee.cause.read().decode('utf
-8'))
340 return self._process_media_selector(media_selection, programme_id)
342 def _process_media_selector(self, media_selection, programme_id):
347 for media in self._extract_medias(media_selection):
348 kind = media.get('kind
')
349 if kind in ('video
', 'audio
'):
350 bitrate = int_or_none(media.get('bitrate
'))
351 encoding = media.get('encoding
')
352 service = media.get('service
')
353 width = int_or_none(media.get('width
'))
354 height = int_or_none(media.get('height
'))
355 file_size = int_or_none(media.get('media_file_size
'))
356 for connection in self._extract_connections(media):
357 href = connection.get('href
')
362 conn_kind = connection.get('kind
')
363 protocol = connection.get('protocol
')
364 supplier = connection.get('supplier
')
365 transfer_format = connection.get('transferFormat
')
366 format_id = supplier or conn_kind or protocol
368 format_id = '%s_%s' % (service, format_id)
370 if supplier == 'asx
':
371 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
374 'format_id
': 'ref
%s_%s' % (i, format_id),
376 elif transfer_format == 'dash
':
377 formats.extend(self._extract_mpd_formats(
378 href, programme_id, mpd_id=format_id, fatal=False))
379 elif transfer_format == 'hls
':
380 formats.extend(self._extract_m3u8_formats(
381 href, programme_id, ext='mp4
', entry_protocol='m3u8_native
',
382 m3u8_id=format_id, fatal=False))
383 if re.search(self._USP_RE, href):
384 usp_formats = self._extract_m3u8_formats(
385 re.sub(self._USP_RE, r'/\
1.ism
/\
1.m3u8
', href),
386 programme_id, ext='mp4
', entry_protocol='m3u8_native
',
387 m3u8_id=format_id, fatal=False)
388 for f in usp_formats:
389 if f.get('height
') and f['height
'] > 720:
392 elif transfer_format == 'hds
':
393 formats.extend(self._extract_f4m_formats(
394 href, programme_id, f4m_id=format_id, fatal=False))
396 if not service and not supplier and bitrate:
397 format_id += '-%d' % bitrate
399 'format_id
': format_id,
400 'filesize
': file_size,
415 if protocol in ('http
', 'https
'):
420 elif protocol == 'rtmp
':
421 application = connection.get('application
', 'ondemand
')
422 auth_string = connection.get('authString
')
423 identifier = connection.get('identifier
')
424 server = connection.get('server
')
426 'url
': '%s://%s/%s?
%s' % (protocol, server, application, auth_string),
427 'play_path
': identifier,
428 'app
': '%s?
%s' % (application, auth_string),
429 'page_url
': 'http
://www
.bbc
.co
.uk
',
430 'player_url
': 'http
://www
.bbc
.co
.uk
/emp
/releases
/iplayer
/revisions
/617463_618125_4/617463_618125_4_emp
.swf
',
437 elif kind == 'captions
':
438 subtitles = self.extract_subtitles(media, programme_id)
439 return formats, subtitles
441 def _download_playlist(self, playlist_id):
443 playlist = self._download_json(
444 'http
://www
.bbc
.co
.uk
/programmes
/%s/playlist
.json
' % playlist_id,
445 playlist_id, 'Downloading playlist JSON
')
447 version = playlist.get('defaultAvailableVersion
')
449 smp_config = version['smpConfig
']
450 title = smp_config['title
']
451 description = smp_config['summary
']
452 for item in smp_config['items
']:
454 if kind not in ('programme
', 'radioProgramme
'):
456 programme_id = item.get('vpid
')
457 duration = int_or_none(item.get('duration
'))
458 formats, subtitles = self._download_media_selector(programme_id)
459 return programme_id, title, description, duration, formats, subtitles
460 except ExtractorError as ee:
461 if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
464 # fallback to legacy playlist
465 return self._process_legacy_playlist(playlist_id)
467 def _process_legacy_playlist_url(self, url, display_id):
468 playlist = self._download_legacy_playlist_url(url, display_id)
469 return self._extract_from_legacy_playlist(playlist, display_id)
471 def _process_legacy_playlist(self, playlist_id):
472 return self._process_legacy_playlist_url(
473 'http
://www
.bbc
.co
.uk
/iplayer
/playlist
/%s' % playlist_id, playlist_id)
475 def _download_legacy_playlist_url(self, url, playlist_id=None):
476 return self._download_xml(
477 url, playlist_id, 'Downloading legacy playlist XML
')
479 def _extract_from_legacy_playlist(self, playlist, playlist_id):
480 no_items = playlist.find('./{%s}noItems
' % self._EMP_PLAYLIST_NS)
481 if no_items is not None:
482 reason = no_items.get('reason
')
483 if reason == 'preAvailability
':
484 msg = 'Episode
%s is not yet available
' % playlist_id
485 elif reason == 'postAvailability
':
486 msg = 'Episode
%s is no longer available
' % playlist_id
487 elif reason == 'noMedia
':
488 msg = 'Episode
%s is not currently available
' % playlist_id
490 msg = 'Episode
%s is not available
: %s' % (playlist_id, reason)
491 raise ExtractorError(msg, expected=True)
493 for item in self._extract_items(playlist):
494 kind = item.get('kind
')
495 if kind not in ('programme
', 'radioProgramme
'):
497 title = playlist.find('./{%s}title
' % self._EMP_PLAYLIST_NS).text
498 description_el = playlist.find('./{%s}summary
' % self._EMP_PLAYLIST_NS)
499 description = description_el.text if description_el is not None else None
501 def get_programme_id(item):
502 def get_from_attributes(item):
503 for p in('identifier
', 'group
'):
505 if value and re.match(r'^
[pb
][\da
-z
]{7}$
', value):
507 get_from_attributes(item)
508 mediator = item.find('./{%s}mediator
' % self._EMP_PLAYLIST_NS)
509 if mediator is not None:
510 return get_from_attributes(mediator)
512 programme_id = get_programme_id(item)
513 duration = int_or_none(item.get('duration
'))
516 formats, subtitles = self._download_media_selector(programme_id)
518 formats, subtitles = self._process_media_selector(item, playlist_id)
519 programme_id = playlist_id
521 return programme_id, title, description, duration, formats, subtitles
523 def _real_extract(self, url):
524 group_id = self._match_id(url)
526 webpage = self._download_webpage(url, group_id, 'Downloading video page
')
528 error = self._search_regex(
529 r'<div
\b[^
>]+\bclass
=["\']smp__message delta["\'][^
>]*>([^
<]+)<',
530 webpage, 'error
', default=None)
532 raise ExtractorError(error, expected=True)
537 tviplayer = self._search_regex(
538 r'mediator\
.bind\
(({.+?}
)\s
*,\s
*document\
.getElementById
',
539 webpage, 'player
', default=None)
542 player = self._parse_json(tviplayer, group_id).get('player
', {})
543 duration = int_or_none(player.get('duration
'))
544 programme_id = player.get('vpid
')
547 programme_id = self._search_regex(
548 r'"vpid"\s
*:\s
*"(%s)"' % self._ID_REGEX, webpage, 'vpid
', fatal=False, default=None)
551 formats, subtitles = self._download_media_selector(programme_id)
552 title = self._og_search_title(webpage, default=None) or self._html_search_regex(
553 (r'<h2
[^
>]+id="parent-title"[^
>]*>(.+?
)</h2
>',
554 r'<div
[^
>]+class="info"[^
>]*>\s
*<h1
>(.+?
)</h1
>'), webpage, 'title
')
555 description = self._search_regex(
556 (r'<p
class="[^"]*medium
-description
[^
"]*">([^
<]+)</p
>',
557 r'<div
[^
>]+class="info_+synopsis"[^
>]*>([^
<]+)</div
>'),
558 webpage, 'description
', default=None)
560 description = self._html_search_meta('description
', webpage)
562 programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
564 self._sort_formats(formats)
569 'description
': description,
570 'thumbnail
': self._og_search_thumbnail(webpage, default=None),
571 'duration
': duration,
573 'subtitles
': subtitles,
577 class BBCIE(BBCCoUkIE):
580 _VALID_URL = r'https?
://(?
:www\
.)?bbc\
.(?
:com|co\
.uk
)/(?
:[^
/]+/)+(?P
<id>[^
/#?]+)'
582 _MEDIASELECTOR_URLS
= [
583 # Provides HQ HLS streams but fails with geolocation in some cases when it's
584 # even not geo restricted at all
585 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
586 # Provides more formats, namely direct mp4 links, but fails on some videos with
587 # notukerror for non UK (?) users (e.g.
588 # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
589 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
590 # Provides fewer formats, but works everywhere for everybody (hopefully)
591 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
595 # article with multiple videos embedded with data-playable containing vpids
596 'url': 'http://www.bbc.com/news/world-europe-32668511',
598 'id': 'world-europe-32668511',
599 'title': 'Russia stages massive WW2 parade despite Western boycott',
600 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
604 # article with multiple videos embedded with data-playable (more videos)
605 'url': 'http://www.bbc.com/news/business-28299555',
607 'id': 'business-28299555',
608 'title': 'Farnborough Airshow: Video highlights',
609 'description': 'BBC reports and video highlights at the Farnborough Airshow.',
614 # article with multiple videos embedded with `new SMP()`
616 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
618 'id': '3662a707-0af9-3149-963f-47bea720b460',
621 'playlist_count': 18,
623 # single video embedded with data-playable containing vpid
624 'url': 'http://www.bbc.com/news/world-europe-32041533',
628 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
629 'description': 'md5:2868290467291b37feda7863f7a83f54',
631 'timestamp': 1427219242,
632 'upload_date': '20150324',
636 'skip_download': True,
639 # article with single video embedded with data-playable containing XML playlist
640 # with direct video links as progressiveDownloadUrl (for now these are extracted)
641 # and playlist with f4m and m3u8 as streamingUrl
642 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
644 'id': '150615_telabyad_kentin_cogu',
646 'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
647 'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
648 'timestamp': 1434397334,
649 'upload_date': '20150615',
652 'skip_download': True,
655 # single video embedded with data-playable containing XML playlists (regional section)
656 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
658 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
660 'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
661 'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
662 'timestamp': 1434713142,
663 'upload_date': '20150619',
666 'skip_download': True,
669 # single video from video playlist embedded with vxp-playlist-data JSON
670 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
674 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
676 'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
679 'skip_download': True,
682 # single video story with digitalData
683 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
687 'title': 'Sri Lanka’s spicy secret',
688 'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
689 'timestamp': 1437674293,
690 'upload_date': '20150723',
694 'skip_download': True,
697 # single video story without digitalData
698 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
702 'title': 'Hyundai Santa Fe Sport: Rock star',
703 'description': 'md5:b042a26142c4154a6e472933cf20793d',
704 'timestamp': 1415867444,
705 'upload_date': '20141113',
709 'skip_download': True,
712 # single video embedded with Morph
713 'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
717 'title': "Nigeria v Japan - Men's First Round",
718 'description': 'Live coverage of the first round from Group B at the Amazonia Arena.',
720 'uploader': 'BBC Sport',
721 'uploader_id': 'bbc_sport',
725 'skip_download': True,
727 'skip': 'Georestricted to UK',
729 # single video with playlist.sxml URL in playlist param
730 'url': 'http://www.bbc.com/sport/0/football/33653409',
734 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
735 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
740 'skip_download': True,
743 # article with multiple videos embedded with playlist.sxml in playlist param
744 'url': 'http://www.bbc.com/sport/0/football/34475836',
747 'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
748 'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
752 # school report article with single video
753 'url': 'http://www.bbc.co.uk/schoolreport/35744779',
756 'title': 'School which breaks down barriers in Jerusalem',
760 # single video with playlist URL from weather section
761 'url': 'http://www.bbc.com/weather/features/33601775',
762 'only_matching': True,
764 # custom redirection to www.bbc.com
765 'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
766 'only_matching': True,
768 # single video article embedded with data-media-vpid
769 'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
770 'only_matching': True,
774 def suitable(cls
, url
):
775 EXCLUDE_IE
= (BBCCoUkIE
, BBCCoUkArticleIE
, BBCCoUkIPlayerPlaylistIE
, BBCCoUkPlaylistIE
)
776 return (False if any(ie
.suitable(url
) for ie
in EXCLUDE_IE
)
777 else super(BBCIE
, cls
).suitable(url
))
779 def _extract_from_media_meta(self
, media_meta
, video_id
):
780 # Direct links to media in media metadata (e.g.
781 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
782 # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
783 source_files
= media_meta
.get('sourceFiles')
787 'format_id': format_id
,
788 'ext': f
.get('encoding'),
789 'tbr': float_or_none(f
.get('bitrate'), 1000),
790 'filesize': int_or_none(f
.get('filesize')),
791 } for format_id
, f
in source_files
.items() if f
.get('url')], []
793 programme_id
= media_meta
.get('externalId')
795 return self
._download
_media
_selector
(programme_id
)
797 # Process playlist.sxml as legacy playlist
798 href
= media_meta
.get('href')
800 playlist
= self
._download
_legacy
_playlist
_url
(href
)
801 _
, _
, _
, _
, formats
, subtitles
= self
._extract
_from
_legacy
_playlist
(playlist
, video_id
)
802 return formats
, subtitles
806 def _extract_from_playlist_sxml(self
, url
, playlist_id
, timestamp
):
807 programme_id
, title
, description
, duration
, formats
, subtitles
= \
808 self
._process
_legacy
_playlist
_url
(url
, playlist_id
)
809 self
._sort
_formats
(formats
)
813 'description': description
,
814 'duration': duration
,
815 'timestamp': timestamp
,
817 'subtitles': subtitles
,
820 def _real_extract(self
, url
):
821 playlist_id
= self
._match
_id
(url
)
823 webpage
= self
._download
_webpage
(url
, playlist_id
)
825 json_ld_info
= self
._search
_json
_ld
(webpage
, playlist_id
, default
={})
826 timestamp
= json_ld_info
.get('timestamp')
828 playlist_title
= json_ld_info
.get('title')
829 if not playlist_title
:
830 playlist_title
= self
._og
_search
_title
(
831 webpage
, default
=None) or self
._html
_search
_regex
(
832 r
'<title>(.+?)</title>', webpage
, 'playlist title', default
=None)
834 playlist_title
= re
.sub(r
'(.+)\s*-\s*BBC.*?$', r
'\1', playlist_title
).strip()
836 playlist_description
= json_ld_info
.get(
837 'description') or self
._og
_search
_description
(webpage
, default
=None)
840 timestamp
= parse_iso8601(self
._search
_regex
(
841 [r
'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
842 r
'itemprop="datePublished"[^>]+datetime="([^"]+)"',
843 r
'"datePublished":\s*"([^"]+)'],
844 webpage
, 'date', default
=None))
848 # article with multiple videos embedded with playlist.sxml (e.g.
849 # http://www.bbc.com/sport/0/football/34475836)
850 playlists
= re
.findall(r
'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage
)
851 playlists
.extend(re
.findall(r
'data-media-id="([^"]+/playlist\.sxml)"', webpage
))
854 self
._extract
_from
_playlist
_sxml
(playlist_url
, playlist_id
, timestamp
)
855 for playlist_url
in playlists
]
857 # news article with multiple videos embedded with data-playable
858 data_playables
= re
.findall(r
'data-playable=(["\'])({.+?}
)\
1', webpage)
860 for _, data_playable_json in data_playables:
861 data_playable = self._parse_json(
862 unescapeHTML(data_playable_json), playlist_id, fatal=False)
863 if not data_playable:
865 settings = data_playable.get('settings
', {})
867 # data-playable with video vpid in settings.playlistObject.items (e.g.
868 # http://www.bbc.com/news/world-us-canada-34473351)
869 playlist_object = settings.get('playlistObject
', {})
871 items = playlist_object.get('items
')
872 if items and isinstance(items, list):
873 title = playlist_object['title
']
874 description = playlist_object.get('summary
')
875 duration = int_or_none(items[0].get('duration
'))
876 programme_id = items[0].get('vpid
')
877 formats, subtitles = self._download_media_selector(programme_id)
878 self._sort_formats(formats)
882 'description
': description,
883 'timestamp
': timestamp,
884 'duration
': duration,
886 'subtitles
': subtitles,
889 # data-playable without vpid but with a playlist.sxml URLs
890 # in otherSettings.playlist (e.g.
891 # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
892 playlist = data_playable.get('otherSettings
', {}).get('playlist', {})
895 for key in ('streaming
', 'progressiveDownload
'):
896 playlist_url = playlist.get('%sUrl
' % key)
900 info = self._extract_from_playlist_sxml(
901 playlist_url, playlist_id, timestamp)
905 entry['title
'] = info['title
']
906 entry['formats
'].extend(info['formats
'])
907 except Exception as e:
908 # Some playlist URL may fail with 500, at the same time
909 # the other one may work fine (e.g.
910 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
911 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
915 self._sort_formats(entry['formats
'])
916 entries.append(entry)
919 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
921 # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
922 programme_id = self._search_regex(
923 [r'data
-(?
:video
-player|media
)-vpid
="(%s)"' % self._ID_REGEX,
924 r'<param
[^
>]+name
="externalIdentifier"[^
>]+value
="(%s)"' % self._ID_REGEX,
925 r'videoId\s
*:\s
*["\'](%s)["\']' % self._ID_REGEX],
926 webpage, 'vpid
', default=None)
929 formats, subtitles = self._download_media_selector(programme_id)
930 self._sort_formats(formats)
931 # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
932 digital_data = self._parse_json(
934 r'var\s
+digitalData\s
*=\s
*({.+?}
);?
\n', webpage, 'digital data
', default='{}'),
935 programme_id, fatal=False)
936 page_info = digital_data.get('page
', {}).get('pageInfo', {})
937 title = page_info.get('pageName
') or self._og_search_title(webpage)
938 description = page_info.get('description
') or self._og_search_description(webpage)
939 timestamp = parse_iso8601(page_info.get('publicationDate
')) or timestamp
943 'description
': description,
944 'timestamp
': timestamp,
946 'subtitles
': subtitles,
949 # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
950 # There are several setPayload calls may be present but the video
951 # seems to be always related to the first one
952 morph_payload = self._parse_json(
954 r'Morph\
.setPayload\
([^
,]+,\s
*({.+?}
)\
);',
955 webpage, 'morph payload
', default='{}'),
956 playlist_id, fatal=False)
958 components = try_get(morph_payload, lambda x: x['body
']['components
'], list) or []
959 for component in components:
960 if not isinstance(component, dict):
962 lead_media = try_get(component, lambda x: x['props
']['leadMedia
'], dict)
965 identifiers = lead_media.get('identifiers
')
966 if not identifiers or not isinstance(identifiers, dict):
968 programme_id = identifiers.get('vpid
') or identifiers.get('playablePid
')
971 title = lead_media.get('title
') or self._og_search_title(webpage)
972 formats, subtitles = self._download_media_selector(programme_id)
973 self._sort_formats(formats)
974 description = lead_media.get('summary
')
975 uploader = lead_media.get('masterBrand
')
976 uploader_id = lead_media.get('mid
')
978 duration_d = lead_media.get('duration
')
979 if isinstance(duration_d, dict):
980 duration = parse_duration(dict_get(
981 duration_d, ('rawDuration
', 'formattedDuration
', 'spokenDuration
')))
985 'description
': description,
986 'duration
': duration,
987 'uploader
': uploader,
988 'uploader_id
': uploader_id,
990 'subtitles
': subtitles,
993 def extract_all(pattern):
994 return list(filter(None, map(
995 lambda s: self._parse_json(s, playlist_id, fatal=False),
996 re.findall(pattern, webpage))))
998 # Multiple video article (e.g.
999 # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
1000 EMBED_URL = r'https?
://(?
:www\
.)?bbc\
.co\
.uk
/(?
:[^
/]+/)+%s(?
:\b[^
"]+)?' % self._ID_REGEX
1002 for match in extract_all(r'new\s+SMP\(({.+?})\)'):
1003 embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
1004 if embed_url and re.match(EMBED_URL, embed_url):
1005 entries.append(embed_url)
1006 entries.extend(re.findall(
1007 r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
1009 return self.playlist_result(
1010 [self.url_result(entry_, 'BBCCoUk') for entry_ in entries],
1011 playlist_id, playlist_title, playlist_description)
1013 # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
1014 medias = extract_all(r"data
-media
-meta
='({[^']+})'")
1017 # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
1018 media_asset = self._search_regex(
1019 r'mediaAssetPage\.init\(\s*({.+?}), "/',
1020 webpage, 'media asset
', default=None)
1022 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
1024 for video in media_asset_page.get('videos
', {}).values():
1025 medias.extend(video.values())
1028 # Multiple video playlist with single `now playing` entry (e.g.
1029 # http://www.bbc.com/news/video_and_audio/must_see/33767813)
1030 vxp_playlist = self._parse_json(
1032 r'<script
[^
>]+class="vxp-playlist-data"[^
>]+type="application/json"[^
>]*>([^
<]+)</script
>',
1033 webpage, 'playlist data
'),
1035 playlist_medias = []
1036 for item in vxp_playlist:
1037 media = item.get('media
')
1040 playlist_medias.append(media)
1041 # Download single video if found media with asset id matching the video id from URL
1042 if item.get('advert
', {}).get('assetId
') == playlist_id:
1045 # Fallback to the whole playlist
1047 medias = playlist_medias
1050 for num, media_meta in enumerate(medias, start=1):
1051 formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
1054 self._sort_formats(formats)
1056 video_id = media_meta.get('externalId
')
1058 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
1060 title = media_meta.get('caption
')
1062 title = playlist_title if len(medias) == 1 else '%s - Video
%s' % (playlist_title, num)
1064 duration = int_or_none(media_meta.get('durationInSeconds
')) or parse_duration(media_meta.get('duration
'))
1067 for image in media_meta.get('images
', {}).values():
1068 images.extend(image.values())
1069 if 'image
' in media_meta:
1070 images.append(media_meta['image
'])
1073 'url
': image.get('href
'),
1074 'width
': int_or_none(image.get('width
')),
1075 'height
': int_or_none(image.get('height
')),
1076 } for image in images]
1081 'thumbnails
': thumbnails,
1082 'duration
': duration,
1083 'timestamp
': timestamp,
1085 'subtitles
': subtitles,
1088 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
1091 class BBCCoUkArticleIE(InfoExtractor):
1092 _VALID_URL = r'https?
://(?
:www\
.)?bbc\
.co\
.uk
/programmes
/articles
/(?P
<id>[a
-zA
-Z0
-9]+)'
1093 IE_NAME = 'bbc
.co
.uk
:article
'
1094 IE_DESC = 'BBC articles
'
1097 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/articles
/3jNQLTMrPlYGTBn0WV6M2MS
/not-your
-typical
-role
-model
-ada
-lovelace
-the
-19th
-century
-programmer
',
1099 'id': '3jNQLTMrPlYGTBn0WV6M2MS
',
1100 'title
': 'Calculating Ada
: The Countess of Computing
- Not your typical role model
: Ada Lovelace the
19th century programmer
- BBC Four
',
1101 'description
': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming
.',
1103 'playlist_count
': 4,
1104 'add_ie
': ['BBCCoUk
'],
1107 def _real_extract(self, url):
1108 playlist_id = self._match_id(url)
1110 webpage = self._download_webpage(url, playlist_id)
1112 title = self._og_search_title(webpage)
1113 description = self._og_search_description(webpage).strip()
1115 entries = [self.url_result(programme_url) for programme_url in re.findall(
1116 r'<div
[^
>]+typeof
="Clip"[^
>]+resource
="([^"]+)"', webpage)]
1118 return self.playlist_result(entries, playlist_id, title, description)
1121 class BBCCoUkPlaylistBaseIE(InfoExtractor):
1122 def _entries(self, webpage, url, playlist_id):
1123 single_page = 'page' in compat_urlparse.parse_qs(
1124 compat_urlparse.urlparse(url).query)
1125 for page_num in itertools.count(2):
1126 for video_id in re.findall(
1127 self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage):
1128 yield self.url_result(
1129 self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
1132 next_page = self._search_regex(
1133 r'<li[^>]+class=(["\'])pagination_
+next\
1[^
>]*><a
[^
>]+href
=(["\'])(?P<url>(?:(?!\2).)+)\2',
1134 webpage, 'next page url', default=None, group='url')
1137 webpage = self._download_webpage(
1138 compat_urlparse.urljoin(url, next_page), playlist_id,
1139 'Downloading page %d' % page_num, page_num)
1141 def _real_extract(self, url):
1142 playlist_id = self._match_id(url)
1144 webpage = self._download_webpage(url, playlist_id)
1146 title, description = self._extract_title_and_description(webpage)
1148 return self.playlist_result(
1149 self._entries(webpage, url, playlist_id),
1150 playlist_id, title, description)
1153 class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
1154 IE_NAME = 'bbc.co.uk:iplayer:playlist'
1155 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
1156 _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
1157 _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
1159 'url
': 'http
://www
.bbc
.co
.uk
/iplayer
/episodes
/b05rcz9v
',
1162 'title
': 'The Disappearance
',
1163 'description
': 'French thriller serial about a missing teenager
.',
1165 'playlist_mincount
': 6,
1166 'skip
': 'This programme
is not currently available on BBC iPlayer
',
1168 # Available for over a year unlike 30 days for most other programmes
1169 'url
': 'http
://www
.bbc
.co
.uk
/iplayer
/group
/p02tcc32
',
1172 'title
': 'Bohemian Icons
',
1173 'description
': 'md5
:683e901041b2fe9ba596f2ab04c4dbe7
',
1175 'playlist_mincount
': 10,
1178 def _extract_title_and_description(self, webpage):
1179 title = self._search_regex(r'<h1
>([^
<]+)</h1
>', webpage, 'title
', fatal=False)
1180 description = self._search_regex(
1181 r'<p
[^
>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
1182 webpage, 'description', fatal=False, group='value')
1183 return title, description
1186 class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
1187 IE_NAME = 'bbc.co.uk:playlist'
1188 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
1189 _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
1190 _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
1192 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b05rcz9v
/clips
',
1195 'title
': 'The Disappearance
- Clips
- BBC Four
',
1196 'description
': 'French thriller serial about a missing teenager
.',
1198 'playlist_mincount
': 7,
1200 # multipage playlist, explicit page
1201 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b00mfl7n
/clips?page
=1',
1204 'title
': 'Frozen Planet
- Clips
- BBC One
',
1205 'description
': 'md5
:65dcbf591ae628dafe32aa6c4a4a0d8c
',
1207 'playlist_mincount
': 24,
1209 # multipage playlist, all pages
1210 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b00mfl7n
/clips
',
1213 'title
': 'Frozen Planet
- Clips
- BBC One
',
1214 'description
': 'md5
:65dcbf591ae628dafe32aa6c4a4a0d8c
',
1216 'playlist_mincount
': 142,
1218 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b05rcz9v
/broadcasts
/2016/06',
1219 'only_matching
': True,
1221 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b05rcz9v
/clips
',
1222 'only_matching
': True,
1224 'url
': 'http
://www
.bbc
.co
.uk
/programmes
/b055jkys
/episodes
/player
',
1225 'only_matching
': True,
1228 def _extract_title_and_description(self, webpage):
1229 title = self._og_search_title(webpage, fatal=False)
1230 description = self._og_search_description(webpage)
1231 return title, description