]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bbc.py
a55a6dbc9dc8e89dda6213650f46e719978cf8ca
[youtubedl] / youtube_dl / extractor / bbc.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..utils import (
8 ExtractorError,
9 float_or_none,
10 int_or_none,
11 parse_duration,
12 parse_iso8601,
13 remove_end,
14 unescapeHTML,
15 )
16 from ..compat import (
17 compat_etree_fromstring,
18 compat_HTTPError,
19 )
20
21
22 class BBCCoUkIE(InfoExtractor):
23 IE_NAME = 'bbc.co.uk'
24 IE_DESC = 'BBC iPlayer'
25 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:programmes/(?!articles/)|iplayer(?:/[^/]+)?/(?:episode/|playlist/))|music/clips[/#])(?P<id>[\da-z]{8})'
26
27 _MEDIASELECTOR_URLS = [
28 # Provides HQ HLS streams with even better quality that pc mediaset but fails
29 # with geolocation in some cases when it's even not geo restricted at all (e.g.
30 # http://www.bbc.co.uk/programmes/b06bp7lf)
31 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
32 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
33 ]
34
35 _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
36 _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
37
38 _NAMESPACES = (
39 _MEDIASELECTION_NS,
40 _EMP_PLAYLIST_NS,
41 )
42
43 _TESTS = [
44 {
45 'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
46 'info_dict': {
47 'id': 'b039d07m',
48 'ext': 'flv',
49 'title': 'Kaleidoscope, Leonard Cohen',
50 'description': 'The Canadian poet and songwriter reflects on his musical career.',
51 'duration': 1740,
52 },
53 'params': {
54 # rtmp download
55 'skip_download': True,
56 }
57 },
58 {
59 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
60 'info_dict': {
61 'id': 'b00yng1d',
62 'ext': 'flv',
63 'title': 'The Man in Black: Series 3: The Printed Name',
64 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
65 'duration': 1800,
66 },
67 'params': {
68 # rtmp download
69 'skip_download': True,
70 },
71 'skip': 'Episode is no longer available on BBC iPlayer Radio',
72 },
73 {
74 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
75 'info_dict': {
76 'id': 'b00yng1d',
77 'ext': 'flv',
78 'title': 'The Voice UK: Series 3: Blind Auditions 5',
79 '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.",
80 'duration': 5100,
81 },
82 'params': {
83 # rtmp download
84 'skip_download': True,
85 },
86 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
87 },
88 {
89 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
90 'info_dict': {
91 'id': 'b03k3pb7',
92 'ext': 'flv',
93 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
94 'description': '2. Invasion',
95 'duration': 3600,
96 },
97 'params': {
98 # rtmp download
99 'skip_download': True,
100 },
101 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
102 }, {
103 'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
104 'info_dict': {
105 'id': 'b04v209v',
106 'ext': 'flv',
107 'title': 'Pete Tong, The Essential New Tune Special',
108 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
109 'duration': 10800,
110 },
111 'params': {
112 # rtmp download
113 'skip_download': True,
114 }
115 }, {
116 'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
117 'note': 'Audio',
118 'info_dict': {
119 'id': 'p02frcch',
120 'ext': 'flv',
121 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
122 'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
123 'duration': 3507,
124 },
125 'params': {
126 # rtmp download
127 'skip_download': True,
128 }
129 }, {
130 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
131 'note': 'Video',
132 'info_dict': {
133 'id': 'p025c103',
134 'ext': 'flv',
135 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
136 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
137 'duration': 226,
138 },
139 'params': {
140 # rtmp download
141 'skip_download': True,
142 }
143 }, {
144 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
145 'info_dict': {
146 'id': 'p02n76xf',
147 'ext': 'flv',
148 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
149 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
150 'duration': 3540,
151 },
152 'params': {
153 # rtmp download
154 'skip_download': True,
155 },
156 'skip': 'geolocation',
157 }, {
158 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
159 'info_dict': {
160 'id': 'b05zmgw1',
161 'ext': 'flv',
162 '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.',
163 'title': 'Royal Academy Summer Exhibition',
164 'duration': 3540,
165 },
166 'params': {
167 # rtmp download
168 'skip_download': True,
169 },
170 'skip': 'geolocation',
171 }, {
172 # iptv-all mediaset fails with geolocation however there is no geo restriction
173 # for this programme at all
174 'url': 'http://www.bbc.co.uk/programmes/b06bp7lf',
175 'info_dict': {
176 'id': 'b06bp7kf',
177 'ext': 'flv',
178 'title': "Annie Mac's Friday Night, B.Traits sits in for Annie",
179 'description': 'B.Traits sits in for Annie Mac with a Mini-Mix from Disclosure.',
180 'duration': 10800,
181 },
182 'params': {
183 # rtmp download
184 'skip_download': True,
185 },
186 }, {
187 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
188 'only_matching': True,
189 }, {
190 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
191 'only_matching': True,
192 }, {
193 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
194 'only_matching': True,
195 }
196 ]
197
198 class MediaSelectionError(Exception):
199 def __init__(self, id):
200 self.id = id
201
202 def _extract_asx_playlist(self, connection, programme_id):
203 asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
204 return [ref.get('href') for ref in asx.findall('./Entry/ref')]
205
206 def _extract_connection(self, connection, programme_id):
207 formats = []
208 kind = connection.get('kind')
209 protocol = connection.get('protocol')
210 supplier = connection.get('supplier')
211 if protocol == 'http':
212 href = connection.get('href')
213 transfer_format = connection.get('transferFormat')
214 # ASX playlist
215 if supplier == 'asx':
216 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
217 formats.append({
218 'url': ref,
219 'format_id': 'ref%s_%s' % (i, supplier),
220 })
221 # Skip DASH until supported
222 elif transfer_format == 'dash':
223 pass
224 elif transfer_format == 'hls':
225 m3u8_formats = self._extract_m3u8_formats(
226 href, programme_id, ext='mp4', entry_protocol='m3u8_native',
227 m3u8_id=supplier, fatal=False)
228 if m3u8_formats:
229 formats.extend(m3u8_formats)
230 # Direct link
231 else:
232 formats.append({
233 'url': href,
234 'format_id': supplier or kind or protocol,
235 })
236 elif protocol == 'rtmp':
237 application = connection.get('application', 'ondemand')
238 auth_string = connection.get('authString')
239 identifier = connection.get('identifier')
240 server = connection.get('server')
241 formats.append({
242 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
243 'play_path': identifier,
244 'app': '%s?%s' % (application, auth_string),
245 'page_url': 'http://www.bbc.co.uk',
246 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
247 'rtmp_live': False,
248 'ext': 'flv',
249 'format_id': supplier,
250 })
251 return formats
252
253 def _extract_items(self, playlist):
254 return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
255
256 def _findall_ns(self, element, xpath):
257 elements = []
258 for ns in self._NAMESPACES:
259 elements.extend(element.findall(xpath % ns))
260 return elements
261
262 def _extract_medias(self, media_selection):
263 error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS)
264 if error is None:
265 media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS)
266 if error is not None:
267 raise BBCCoUkIE.MediaSelectionError(error.get('id'))
268 return self._findall_ns(media_selection, './{%s}media')
269
270 def _extract_connections(self, media):
271 return self._findall_ns(media, './{%s}connection')
272
273 def _extract_video(self, media, programme_id):
274 formats = []
275 vbr = int_or_none(media.get('bitrate'))
276 vcodec = media.get('encoding')
277 service = media.get('service')
278 width = int_or_none(media.get('width'))
279 height = int_or_none(media.get('height'))
280 file_size = int_or_none(media.get('media_file_size'))
281 for connection in self._extract_connections(media):
282 conn_formats = self._extract_connection(connection, programme_id)
283 for format in conn_formats:
284 format.update({
285 'width': width,
286 'height': height,
287 'vbr': vbr,
288 'vcodec': vcodec,
289 'filesize': file_size,
290 })
291 if service:
292 format['format_id'] = '%s_%s' % (service, format['format_id'])
293 formats.extend(conn_formats)
294 return formats
295
296 def _extract_audio(self, media, programme_id):
297 formats = []
298 abr = int_or_none(media.get('bitrate'))
299 acodec = media.get('encoding')
300 service = media.get('service')
301 for connection in self._extract_connections(media):
302 conn_formats = self._extract_connection(connection, programme_id)
303 for format in conn_formats:
304 format.update({
305 'format_id': '%s_%s' % (service, format['format_id']),
306 'abr': abr,
307 'acodec': acodec,
308 })
309 formats.extend(conn_formats)
310 return formats
311
312 def _get_subtitles(self, media, programme_id):
313 subtitles = {}
314 for connection in self._extract_connections(media):
315 captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
316 lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
317 subtitles[lang] = [
318 {
319 'url': connection.get('href'),
320 'ext': 'ttml',
321 },
322 ]
323 return subtitles
324
325 def _raise_extractor_error(self, media_selection_error):
326 raise ExtractorError(
327 '%s returned error: %s' % (self.IE_NAME, media_selection_error.id),
328 expected=True)
329
330 def _download_media_selector(self, programme_id):
331 last_exception = None
332 for mediaselector_url in self._MEDIASELECTOR_URLS:
333 try:
334 return self._download_media_selector_url(
335 mediaselector_url % programme_id, programme_id)
336 except BBCCoUkIE.MediaSelectionError as e:
337 if e.id in ('notukerror', 'geolocation'):
338 last_exception = e
339 continue
340 self._raise_extractor_error(e)
341 self._raise_extractor_error(last_exception)
342
343 def _download_media_selector_url(self, url, programme_id=None):
344 try:
345 media_selection = self._download_xml(
346 url, programme_id, 'Downloading media selection XML')
347 except ExtractorError as ee:
348 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
349 media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8'))
350 else:
351 raise
352 return self._process_media_selector(media_selection, programme_id)
353
354 def _process_media_selector(self, media_selection, programme_id):
355 formats = []
356 subtitles = None
357
358 for media in self._extract_medias(media_selection):
359 kind = media.get('kind')
360 if kind == 'audio':
361 formats.extend(self._extract_audio(media, programme_id))
362 elif kind == 'video':
363 formats.extend(self._extract_video(media, programme_id))
364 elif kind == 'captions':
365 subtitles = self.extract_subtitles(media, programme_id)
366 return formats, subtitles
367
368 def _download_playlist(self, playlist_id):
369 try:
370 playlist = self._download_json(
371 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
372 playlist_id, 'Downloading playlist JSON')
373
374 version = playlist.get('defaultAvailableVersion')
375 if version:
376 smp_config = version['smpConfig']
377 title = smp_config['title']
378 description = smp_config['summary']
379 for item in smp_config['items']:
380 kind = item['kind']
381 if kind != 'programme' and kind != 'radioProgramme':
382 continue
383 programme_id = item.get('vpid')
384 duration = int_or_none(item.get('duration'))
385 formats, subtitles = self._download_media_selector(programme_id)
386 return programme_id, title, description, duration, formats, subtitles
387 except ExtractorError as ee:
388 if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
389 raise
390
391 # fallback to legacy playlist
392 return self._process_legacy_playlist(playlist_id)
393
394 def _process_legacy_playlist_url(self, url, display_id):
395 playlist = self._download_legacy_playlist_url(url, display_id)
396 return self._extract_from_legacy_playlist(playlist, display_id)
397
398 def _process_legacy_playlist(self, playlist_id):
399 return self._process_legacy_playlist_url(
400 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
401
402 def _download_legacy_playlist_url(self, url, playlist_id=None):
403 return self._download_xml(
404 url, playlist_id, 'Downloading legacy playlist XML')
405
406 def _extract_from_legacy_playlist(self, playlist, playlist_id):
407 no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS)
408 if no_items is not None:
409 reason = no_items.get('reason')
410 if reason == 'preAvailability':
411 msg = 'Episode %s is not yet available' % playlist_id
412 elif reason == 'postAvailability':
413 msg = 'Episode %s is no longer available' % playlist_id
414 elif reason == 'noMedia':
415 msg = 'Episode %s is not currently available' % playlist_id
416 else:
417 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
418 raise ExtractorError(msg, expected=True)
419
420 for item in self._extract_items(playlist):
421 kind = item.get('kind')
422 if kind != 'programme' and kind != 'radioProgramme':
423 continue
424 title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
425 description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
426 description = description_el.text if description_el is not None else None
427
428 def get_programme_id(item):
429 def get_from_attributes(item):
430 for p in('identifier', 'group'):
431 value = item.get(p)
432 if value and re.match(r'^[pb][\da-z]{7}$', value):
433 return value
434 get_from_attributes(item)
435 mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS)
436 if mediator is not None:
437 return get_from_attributes(mediator)
438
439 programme_id = get_programme_id(item)
440 duration = int_or_none(item.get('duration'))
441
442 if programme_id:
443 formats, subtitles = self._download_media_selector(programme_id)
444 else:
445 formats, subtitles = self._process_media_selector(item, playlist_id)
446 programme_id = playlist_id
447
448 return programme_id, title, description, duration, formats, subtitles
449
450 def _real_extract(self, url):
451 group_id = self._match_id(url)
452
453 webpage = self._download_webpage(url, group_id, 'Downloading video page')
454
455 programme_id = None
456
457 tviplayer = self._search_regex(
458 r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
459 webpage, 'player', default=None)
460
461 if tviplayer:
462 player = self._parse_json(tviplayer, group_id).get('player', {})
463 duration = int_or_none(player.get('duration'))
464 programme_id = player.get('vpid')
465
466 if not programme_id:
467 programme_id = self._search_regex(
468 r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
469
470 if programme_id:
471 formats, subtitles = self._download_media_selector(programme_id)
472 title = self._og_search_title(webpage)
473 description = self._search_regex(
474 r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
475 webpage, 'description', fatal=False)
476 else:
477 programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
478
479 self._sort_formats(formats)
480
481 return {
482 'id': programme_id,
483 'title': title,
484 'description': description,
485 'thumbnail': self._og_search_thumbnail(webpage, default=None),
486 'duration': duration,
487 'formats': formats,
488 'subtitles': subtitles,
489 }
490
491
492 class BBCIE(BBCCoUkIE):
493 IE_NAME = 'bbc'
494 IE_DESC = 'BBC'
495 _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
496
497 _MEDIASELECTOR_URLS = [
498 # Provides HQ HLS streams but fails with geolocation in some cases when it's
499 # even not geo restricted at all
500 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
501 # Provides more formats, namely direct mp4 links, but fails on some videos with
502 # notukerror for non UK (?) users (e.g.
503 # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
504 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
505 # Provides fewer formats, but works everywhere for everybody (hopefully)
506 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
507 ]
508
509 _TESTS = [{
510 # article with multiple videos embedded with data-playable containing vpids
511 'url': 'http://www.bbc.com/news/world-europe-32668511',
512 'info_dict': {
513 'id': 'world-europe-32668511',
514 'title': 'Russia stages massive WW2 parade despite Western boycott',
515 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
516 },
517 'playlist_count': 2,
518 }, {
519 # article with multiple videos embedded with data-playable (more videos)
520 'url': 'http://www.bbc.com/news/business-28299555',
521 'info_dict': {
522 'id': 'business-28299555',
523 'title': 'Farnborough Airshow: Video highlights',
524 'description': 'BBC reports and video highlights at the Farnborough Airshow.',
525 },
526 'playlist_count': 9,
527 'skip': 'Save time',
528 }, {
529 # article with multiple videos embedded with `new SMP()`
530 # broken
531 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
532 'info_dict': {
533 'id': '3662a707-0af9-3149-963f-47bea720b460',
534 'title': 'BBC Blogs - Adam Curtis - BUGGER',
535 },
536 'playlist_count': 18,
537 }, {
538 # single video embedded with data-playable containing vpid
539 'url': 'http://www.bbc.com/news/world-europe-32041533',
540 'info_dict': {
541 'id': 'p02mprgb',
542 'ext': 'mp4',
543 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
544 'description': 'md5:2868290467291b37feda7863f7a83f54',
545 'duration': 47,
546 'timestamp': 1427219242,
547 'upload_date': '20150324',
548 },
549 'params': {
550 # rtmp download
551 'skip_download': True,
552 }
553 }, {
554 # article with single video embedded with data-playable containing XML playlist
555 # with direct video links as progressiveDownloadUrl (for now these are extracted)
556 # and playlist with f4m and m3u8 as streamingUrl
557 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
558 'info_dict': {
559 'id': '150615_telabyad_kentin_cogu',
560 'ext': 'mp4',
561 'title': "YPG: Tel Abyad'ın tamamı kontrolĆ¼mĆ¼zde",
562 'timestamp': 1434397334,
563 'upload_date': '20150615',
564 },
565 'params': {
566 'skip_download': True,
567 }
568 }, {
569 # single video embedded with data-playable containing XML playlists (regional section)
570 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
571 'info_dict': {
572 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
573 'ext': 'mp4',
574 'title': 'Honduras militariza sus hospitales por nuevo escĆ”ndalo de corrupciĆ³n',
575 'timestamp': 1434713142,
576 'upload_date': '20150619',
577 },
578 'params': {
579 'skip_download': True,
580 }
581 }, {
582 # single video from video playlist embedded with vxp-playlist-data JSON
583 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
584 'info_dict': {
585 'id': 'p02w6qjc',
586 'ext': 'mp4',
587 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
588 'duration': 56,
589 },
590 'params': {
591 'skip_download': True,
592 }
593 }, {
594 # single video story with digitalData
595 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
596 'info_dict': {
597 'id': 'p02q6gc4',
598 'ext': 'flv',
599 'title': 'Sri Lankaā€™s spicy secret',
600 'description': 'As a new train line to Jaffna opens up the countryā€™s north, travellers can experience a truly distinct slice of Tamil culture.',
601 'timestamp': 1437674293,
602 'upload_date': '20150723',
603 },
604 'params': {
605 # rtmp download
606 'skip_download': True,
607 }
608 }, {
609 # single video story without digitalData
610 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
611 'info_dict': {
612 'id': 'p018zqqg',
613 'ext': 'mp4',
614 'title': 'Hyundai Santa Fe Sport: Rock star',
615 'description': 'md5:b042a26142c4154a6e472933cf20793d',
616 'timestamp': 1415867444,
617 'upload_date': '20141113',
618 },
619 'params': {
620 # rtmp download
621 'skip_download': True,
622 }
623 }, {
624 # single video with playlist.sxml URL in playlist param
625 'url': 'http://www.bbc.com/sport/0/football/33653409',
626 'info_dict': {
627 'id': 'p02xycnp',
628 'ext': 'mp4',
629 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
630 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
631 'duration': 140,
632 },
633 'params': {
634 # rtmp download
635 'skip_download': True,
636 }
637 }, {
638 # article with multiple videos embedded with playlist.sxml in playlist param
639 'url': 'http://www.bbc.com/sport/0/football/34475836',
640 'info_dict': {
641 'id': '34475836',
642 'title': 'What Liverpool can expect from Klopp',
643 },
644 'playlist_count': 3,
645 }, {
646 # single video with playlist URL from weather section
647 'url': 'http://www.bbc.com/weather/features/33601775',
648 'only_matching': True,
649 }, {
650 # custom redirection to www.bbc.com
651 'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
652 'only_matching': True,
653 }]
654
655 @classmethod
656 def suitable(cls, url):
657 return False if BBCCoUkIE.suitable(url) or BBCCoUkArticleIE.suitable(url) else super(BBCIE, cls).suitable(url)
658
659 def _extract_from_media_meta(self, media_meta, video_id):
660 # Direct links to media in media metadata (e.g.
661 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
662 # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
663 source_files = media_meta.get('sourceFiles')
664 if source_files:
665 return [{
666 'url': f['url'],
667 'format_id': format_id,
668 'ext': f.get('encoding'),
669 'tbr': float_or_none(f.get('bitrate'), 1000),
670 'filesize': int_or_none(f.get('filesize')),
671 } for format_id, f in source_files.items() if f.get('url')], []
672
673 programme_id = media_meta.get('externalId')
674 if programme_id:
675 return self._download_media_selector(programme_id)
676
677 # Process playlist.sxml as legacy playlist
678 href = media_meta.get('href')
679 if href:
680 playlist = self._download_legacy_playlist_url(href)
681 _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
682 return formats, subtitles
683
684 return [], []
685
686 def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
687 programme_id, title, description, duration, formats, subtitles = \
688 self._process_legacy_playlist_url(url, playlist_id)
689 self._sort_formats(formats)
690 return {
691 'id': programme_id,
692 'title': title,
693 'description': description,
694 'duration': duration,
695 'timestamp': timestamp,
696 'formats': formats,
697 'subtitles': subtitles,
698 }
699
700 def _real_extract(self, url):
701 playlist_id = self._match_id(url)
702
703 webpage = self._download_webpage(url, playlist_id)
704
705 timestamp = None
706 playlist_title = None
707 playlist_description = None
708
709 ld = self._parse_json(
710 self._search_regex(
711 r'(?s)<script type="application/ld\+json">(.+?)</script>',
712 webpage, 'ld json', default='{}'),
713 playlist_id, fatal=False)
714 if ld:
715 timestamp = parse_iso8601(ld.get('datePublished'))
716 playlist_title = ld.get('headline')
717 playlist_description = ld.get('articleBody')
718
719 if not timestamp:
720 timestamp = parse_iso8601(self._search_regex(
721 [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
722 r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
723 r'"datePublished":\s*"([^"]+)'],
724 webpage, 'date', default=None))
725
726 entries = []
727
728 # article with multiple videos embedded with playlist.sxml (e.g.
729 # http://www.bbc.com/sport/0/football/34475836)
730 playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
731 if playlists:
732 entries = [
733 self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
734 for playlist_url in playlists]
735
736 # news article with multiple videos embedded with data-playable
737 data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
738 if data_playables:
739 for _, data_playable_json in data_playables:
740 data_playable = self._parse_json(
741 unescapeHTML(data_playable_json), playlist_id, fatal=False)
742 if not data_playable:
743 continue
744 settings = data_playable.get('settings', {})
745 if settings:
746 # data-playable with video vpid in settings.playlistObject.items (e.g.
747 # http://www.bbc.com/news/world-us-canada-34473351)
748 playlist_object = settings.get('playlistObject', {})
749 if playlist_object:
750 items = playlist_object.get('items')
751 if items and isinstance(items, list):
752 title = playlist_object['title']
753 description = playlist_object.get('summary')
754 duration = int_or_none(items[0].get('duration'))
755 programme_id = items[0].get('vpid')
756 formats, subtitles = self._download_media_selector(programme_id)
757 self._sort_formats(formats)
758 entries.append({
759 'id': programme_id,
760 'title': title,
761 'description': description,
762 'timestamp': timestamp,
763 'duration': duration,
764 'formats': formats,
765 'subtitles': subtitles,
766 })
767 else:
768 # data-playable without vpid but with a playlist.sxml URLs
769 # in otherSettings.playlist (e.g.
770 # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
771 playlist = data_playable.get('otherSettings', {}).get('playlist', {})
772 if playlist:
773 entries.append(self._extract_from_playlist_sxml(
774 playlist.get('progressiveDownloadUrl'), playlist_id, timestamp))
775
776 if entries:
777 playlist_title = playlist_title or remove_end(self._og_search_title(webpage), ' - BBC News')
778 playlist_description = playlist_description or self._og_search_description(webpage, default=None)
779 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
780
781 # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
782 programme_id = self._search_regex(
783 [r'data-video-player-vpid="([\da-z]{8})"',
784 r'<param[^>]+name="externalIdentifier"[^>]+value="([\da-z]{8})"'],
785 webpage, 'vpid', default=None)
786
787 if programme_id:
788 formats, subtitles = self._download_media_selector(programme_id)
789 self._sort_formats(formats)
790 # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
791 digital_data = self._parse_json(
792 self._search_regex(
793 r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
794 programme_id, fatal=False)
795 page_info = digital_data.get('page', {}).get('pageInfo', {})
796 title = page_info.get('pageName') or self._og_search_title(webpage)
797 description = page_info.get('description') or self._og_search_description(webpage)
798 timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
799 return {
800 'id': programme_id,
801 'title': title,
802 'description': description,
803 'timestamp': timestamp,
804 'formats': formats,
805 'subtitles': subtitles,
806 }
807
808 playlist_title = self._html_search_regex(
809 r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'playlist title')
810 playlist_description = self._og_search_description(webpage, default=None)
811
812 def extract_all(pattern):
813 return list(filter(None, map(
814 lambda s: self._parse_json(s, playlist_id, fatal=False),
815 re.findall(pattern, webpage))))
816
817 # Multiple video article (e.g.
818 # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
819 EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+[\da-z]{8}(?:\b[^"]+)?'
820 entries = []
821 for match in extract_all(r'new\s+SMP\(({.+?})\)'):
822 embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
823 if embed_url and re.match(EMBED_URL, embed_url):
824 entries.append(embed_url)
825 entries.extend(re.findall(
826 r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
827 if entries:
828 return self.playlist_result(
829 [self.url_result(entry, 'BBCCoUk') for entry in entries],
830 playlist_id, playlist_title, playlist_description)
831
832 # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
833 medias = extract_all(r"data-media-meta='({[^']+})'")
834
835 if not medias:
836 # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
837 media_asset = self._search_regex(
838 r'mediaAssetPage\.init\(\s*({.+?}), "/',
839 webpage, 'media asset', default=None)
840 if media_asset:
841 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
842 medias = []
843 for video in media_asset_page.get('videos', {}).values():
844 medias.extend(video.values())
845
846 if not medias:
847 # Multiple video playlist with single `now playing` entry (e.g.
848 # http://www.bbc.com/news/video_and_audio/must_see/33767813)
849 vxp_playlist = self._parse_json(
850 self._search_regex(
851 r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>',
852 webpage, 'playlist data'),
853 playlist_id)
854 playlist_medias = []
855 for item in vxp_playlist:
856 media = item.get('media')
857 if not media:
858 continue
859 playlist_medias.append(media)
860 # Download single video if found media with asset id matching the video id from URL
861 if item.get('advert', {}).get('assetId') == playlist_id:
862 medias = [media]
863 break
864 # Fallback to the whole playlist
865 if not medias:
866 medias = playlist_medias
867
868 entries = []
869 for num, media_meta in enumerate(medias, start=1):
870 formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
871 if not formats:
872 continue
873 self._sort_formats(formats)
874
875 video_id = media_meta.get('externalId')
876 if not video_id:
877 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
878
879 title = media_meta.get('caption')
880 if not title:
881 title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
882
883 duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
884
885 images = []
886 for image in media_meta.get('images', {}).values():
887 images.extend(image.values())
888 if 'image' in media_meta:
889 images.append(media_meta['image'])
890
891 thumbnails = [{
892 'url': image.get('href'),
893 'width': int_or_none(image.get('width')),
894 'height': int_or_none(image.get('height')),
895 } for image in images]
896
897 entries.append({
898 'id': video_id,
899 'title': title,
900 'thumbnails': thumbnails,
901 'duration': duration,
902 'timestamp': timestamp,
903 'formats': formats,
904 'subtitles': subtitles,
905 })
906
907 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
908
909
910 class BBCCoUkArticleIE(InfoExtractor):
911 _VALID_URL = 'http://www.bbc.co.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
912 IE_NAME = 'bbc.co.uk:article'
913 IE_DESC = 'BBC articles'
914
915 _TEST = {
916 'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer',
917 'info_dict': {
918 'id': '3jNQLTMrPlYGTBn0WV6M2MS',
919 'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four',
920 'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.',
921 },
922 'playlist_count': 4,
923 'add_ie': ['BBCCoUk'],
924 }
925
926 def _real_extract(self, url):
927 playlist_id = self._match_id(url)
928
929 webpage = self._download_webpage(url, playlist_id)
930
931 title = self._og_search_title(webpage)
932 description = self._og_search_description(webpage).strip()
933
934 entries = [self.url_result(programme_url) for programme_url in re.findall(
935 r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
936
937 return self.playlist_result(entries, playlist_id, title, description)