]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bbc.py
dd65b8d86a36f3bbb486072260f14ab652a1868e
[youtubedl] / youtube_dl / extractor / bbc.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5 import itertools
6
7 from .common import InfoExtractor
8 from ..utils import (
9 dict_get,
10 ExtractorError,
11 float_or_none,
12 int_or_none,
13 parse_duration,
14 parse_iso8601,
15 try_get,
16 unescapeHTML,
17 )
18 from ..compat import (
19 compat_etree_fromstring,
20 compat_HTTPError,
21 compat_urlparse,
22 )
23
24
25 class BBCCoUkIE(InfoExtractor):
26 IE_NAME = 'bbc.co.uk'
27 IE_DESC = 'BBC iPlayer'
28 _ID_REGEX = r'[pb][\da-z]{7}'
29 _VALID_URL = r'''(?x)
30 https?://
31 (?:www\.)?bbc\.co\.uk/
32 (?:
33 programmes/(?!articles/)|
34 iplayer(?:/[^/]+)?/(?:episode/|playlist/)|
35 music/clips[/#]|
36 radio/player/
37 )
38 (?P<id>%s)(?!/(?:episodes|broadcasts|clips))
39 ''' % _ID_REGEX
40
41 _MEDIASELECTOR_URLS = [
42 # Provides HQ HLS streams with even better quality that pc mediaset but fails
43 # with geolocation in some cases when it's even not geo restricted at all (e.g.
44 # http://www.bbc.co.uk/programmes/b06bp7lf). Also may fail with selectionunavailable.
45 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
46 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
47 ]
48
49 _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
50 _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
51
52 _NAMESPACES = (
53 _MEDIASELECTION_NS,
54 _EMP_PLAYLIST_NS,
55 )
56
57 _TESTS = [
58 {
59 'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
60 'info_dict': {
61 'id': 'b039d07m',
62 'ext': 'flv',
63 'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
64 'description': 'The Canadian poet and songwriter reflects on his musical career.',
65 },
66 'params': {
67 # rtmp download
68 'skip_download': True,
69 }
70 },
71 {
72 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
73 'info_dict': {
74 'id': 'b00yng1d',
75 'ext': 'flv',
76 'title': 'The Man in Black: Series 3: The Printed Name',
77 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
78 'duration': 1800,
79 },
80 'params': {
81 # rtmp download
82 'skip_download': True,
83 },
84 'skip': 'Episode is no longer available on BBC iPlayer Radio',
85 },
86 {
87 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
88 'info_dict': {
89 'id': 'b00yng1d',
90 'ext': 'flv',
91 'title': 'The Voice UK: Series 3: Blind Auditions 5',
92 'description': 'Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.',
93 'duration': 5100,
94 },
95 'params': {
96 # rtmp download
97 'skip_download': True,
98 },
99 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
100 },
101 {
102 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
103 'info_dict': {
104 'id': 'b03k3pb7',
105 'ext': 'flv',
106 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
107 'description': '2. Invasion',
108 'duration': 3600,
109 },
110 'params': {
111 # rtmp download
112 'skip_download': True,
113 },
114 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
115 }, {
116 'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
117 'info_dict': {
118 'id': 'b04v209v',
119 'ext': 'flv',
120 'title': 'Pete Tong, The Essential New Tune Special',
121 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
122 'duration': 10800,
123 },
124 'params': {
125 # rtmp download
126 'skip_download': True,
127 },
128 'skip': 'Episode is no longer available on BBC iPlayer Radio',
129 }, {
130 'url': 'http://www.bbc.co.uk/music/clips/p022h44b',
131 'note': 'Audio',
132 'info_dict': {
133 'id': 'p022h44j',
134 'ext': 'flv',
135 'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
136 'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
137 'duration': 227,
138 },
139 'params': {
140 # rtmp download
141 'skip_download': True,
142 }
143 }, {
144 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
145 'note': 'Video',
146 'info_dict': {
147 'id': 'p025c103',
148 'ext': 'flv',
149 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
150 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
151 'duration': 226,
152 },
153 'params': {
154 # rtmp download
155 'skip_download': True,
156 }
157 }, {
158 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
159 'info_dict': {
160 'id': 'p02n76xf',
161 'ext': 'flv',
162 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
163 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
164 'duration': 3540,
165 },
166 'params': {
167 # rtmp download
168 'skip_download': True,
169 },
170 'skip': 'geolocation',
171 }, {
172 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
173 'info_dict': {
174 'id': 'b05zmgw1',
175 'ext': 'flv',
176 '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.',
177 'title': 'Royal Academy Summer Exhibition',
178 'duration': 3540,
179 },
180 'params': {
181 # rtmp download
182 'skip_download': True,
183 },
184 'skip': 'geolocation',
185 }, {
186 # iptv-all mediaset fails with geolocation however there is no geo restriction
187 # for this programme at all
188 'url': 'http://www.bbc.co.uk/programmes/b06rkn85',
189 'info_dict': {
190 'id': 'b06rkms3',
191 'ext': 'flv',
192 'title': "Best of the Mini-Mixes 2015: Part 3, Annie Mac's Friday Night - BBC Radio 1",
193 'description': "Annie has part three in the Best of the Mini-Mixes 2015, plus the year's Most Played!",
194 },
195 'params': {
196 # rtmp download
197 'skip_download': True,
198 },
199 'skip': 'Now it\'s really geo-restricted',
200 }, {
201 # compact player (https://github.com/rg3/youtube-dl/issues/8147)
202 'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
203 'info_dict': {
204 'id': 'p028bfkj',
205 'ext': 'flv',
206 'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
207 'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
208 },
209 'params': {
210 # rtmp download
211 'skip_download': True,
212 },
213 }, {
214 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
215 'only_matching': True,
216 }, {
217 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
218 'only_matching': True,
219 }, {
220 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
221 'only_matching': True,
222 }, {
223 'url': 'http://www.bbc.co.uk/radio/player/p03cchwf',
224 'only_matching': True,
225 }
226 ]
227
228 _USP_RE = r'/([^/]+?)\.ism(?:\.hlsv2\.ism)?/[^/]+\.m3u8'
229
230 class MediaSelectionError(Exception):
231 def __init__(self, id):
232 self.id = id
233
234 def _extract_asx_playlist(self, connection, programme_id):
235 asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
236 return [ref.get('href') for ref in asx.findall('./Entry/ref')]
237
238 def _extract_items(self, playlist):
239 return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
240
241 def _findall_ns(self, element, xpath):
242 elements = []
243 for ns in self._NAMESPACES:
244 elements.extend(element.findall(xpath % ns))
245 return elements
246
247 def _extract_medias(self, media_selection):
248 error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS)
249 if error is None:
250 media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS)
251 if error is not None:
252 raise BBCCoUkIE.MediaSelectionError(error.get('id'))
253 return self._findall_ns(media_selection, './{%s}media')
254
255 def _extract_connections(self, media):
256 return self._findall_ns(media, './{%s}connection')
257
258 def _get_subtitles(self, media, programme_id):
259 subtitles = {}
260 for connection in self._extract_connections(media):
261 captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
262 lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
263 subtitles[lang] = [
264 {
265 'url': connection.get('href'),
266 'ext': 'ttml',
267 },
268 ]
269 return subtitles
270
271 def _raise_extractor_error(self, media_selection_error):
272 raise ExtractorError(
273 '%s returned error: %s' % (self.IE_NAME, media_selection_error.id),
274 expected=True)
275
276 def _download_media_selector(self, programme_id):
277 last_exception = None
278 for mediaselector_url in self._MEDIASELECTOR_URLS:
279 try:
280 return self._download_media_selector_url(
281 mediaselector_url % programme_id, programme_id)
282 except BBCCoUkIE.MediaSelectionError as e:
283 if e.id in ('notukerror', 'geolocation', 'selectionunavailable'):
284 last_exception = e
285 continue
286 self._raise_extractor_error(e)
287 self._raise_extractor_error(last_exception)
288
289 def _download_media_selector_url(self, url, programme_id=None):
290 try:
291 media_selection = self._download_xml(
292 url, programme_id, 'Downloading media selection XML')
293 except ExtractorError as ee:
294 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code in (403, 404):
295 media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8'))
296 else:
297 raise
298 return self._process_media_selector(media_selection, programme_id)
299
300 def _process_media_selector(self, media_selection, programme_id):
301 formats = []
302 subtitles = None
303 urls = []
304
305 for media in self._extract_medias(media_selection):
306 kind = media.get('kind')
307 if kind in ('video', 'audio'):
308 bitrate = int_or_none(media.get('bitrate'))
309 encoding = media.get('encoding')
310 service = media.get('service')
311 width = int_or_none(media.get('width'))
312 height = int_or_none(media.get('height'))
313 file_size = int_or_none(media.get('media_file_size'))
314 for connection in self._extract_connections(media):
315 href = connection.get('href')
316 if href in urls:
317 continue
318 if href:
319 urls.append(href)
320 conn_kind = connection.get('kind')
321 protocol = connection.get('protocol')
322 supplier = connection.get('supplier')
323 transfer_format = connection.get('transferFormat')
324 format_id = supplier or conn_kind or protocol
325 if service:
326 format_id = '%s_%s' % (service, format_id)
327 # ASX playlist
328 if supplier == 'asx':
329 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
330 formats.append({
331 'url': ref,
332 'format_id': 'ref%s_%s' % (i, format_id),
333 })
334 elif transfer_format == 'dash':
335 formats.extend(self._extract_mpd_formats(
336 href, programme_id, mpd_id=format_id, fatal=False))
337 elif transfer_format == 'hls':
338 formats.extend(self._extract_m3u8_formats(
339 href, programme_id, ext='mp4', entry_protocol='m3u8_native',
340 m3u8_id=format_id, fatal=False))
341 if re.search(self._USP_RE, href):
342 usp_formats = self._extract_m3u8_formats(
343 re.sub(self._USP_RE, r'/\1.ism/\1.m3u8', href),
344 programme_id, ext='mp4', entry_protocol='m3u8_native',
345 m3u8_id=format_id, fatal=False)
346 for f in usp_formats:
347 if f.get('height') and f['height'] > 720:
348 continue
349 formats.append(f)
350 elif transfer_format == 'hds':
351 formats.extend(self._extract_f4m_formats(
352 href, programme_id, f4m_id=format_id, fatal=False))
353 else:
354 if not service and not supplier and bitrate:
355 format_id += '-%d' % bitrate
356 fmt = {
357 'format_id': format_id,
358 'filesize': file_size,
359 }
360 if kind == 'video':
361 fmt.update({
362 'width': width,
363 'height': height,
364 'tbr': bitrate,
365 'vcodec': encoding,
366 })
367 else:
368 fmt.update({
369 'abr': bitrate,
370 'acodec': encoding,
371 'vcodec': 'none',
372 })
373 if protocol in ('http', 'https'):
374 # Direct link
375 fmt.update({
376 'url': href,
377 })
378 elif protocol == 'rtmp':
379 application = connection.get('application', 'ondemand')
380 auth_string = connection.get('authString')
381 identifier = connection.get('identifier')
382 server = connection.get('server')
383 fmt.update({
384 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
385 'play_path': identifier,
386 'app': '%s?%s' % (application, auth_string),
387 'page_url': 'http://www.bbc.co.uk',
388 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
389 'rtmp_live': False,
390 'ext': 'flv',
391 })
392 else:
393 continue
394 formats.append(fmt)
395 elif kind == 'captions':
396 subtitles = self.extract_subtitles(media, programme_id)
397 return formats, subtitles
398
399 def _download_playlist(self, playlist_id):
400 try:
401 playlist = self._download_json(
402 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
403 playlist_id, 'Downloading playlist JSON')
404
405 version = playlist.get('defaultAvailableVersion')
406 if version:
407 smp_config = version['smpConfig']
408 title = smp_config['title']
409 description = smp_config['summary']
410 for item in smp_config['items']:
411 kind = item['kind']
412 if kind not in ('programme', 'radioProgramme'):
413 continue
414 programme_id = item.get('vpid')
415 duration = int_or_none(item.get('duration'))
416 formats, subtitles = self._download_media_selector(programme_id)
417 return programme_id, title, description, duration, formats, subtitles
418 except ExtractorError as ee:
419 if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
420 raise
421
422 # fallback to legacy playlist
423 return self._process_legacy_playlist(playlist_id)
424
425 def _process_legacy_playlist_url(self, url, display_id):
426 playlist = self._download_legacy_playlist_url(url, display_id)
427 return self._extract_from_legacy_playlist(playlist, display_id)
428
429 def _process_legacy_playlist(self, playlist_id):
430 return self._process_legacy_playlist_url(
431 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
432
433 def _download_legacy_playlist_url(self, url, playlist_id=None):
434 return self._download_xml(
435 url, playlist_id, 'Downloading legacy playlist XML')
436
437 def _extract_from_legacy_playlist(self, playlist, playlist_id):
438 no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS)
439 if no_items is not None:
440 reason = no_items.get('reason')
441 if reason == 'preAvailability':
442 msg = 'Episode %s is not yet available' % playlist_id
443 elif reason == 'postAvailability':
444 msg = 'Episode %s is no longer available' % playlist_id
445 elif reason == 'noMedia':
446 msg = 'Episode %s is not currently available' % playlist_id
447 else:
448 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
449 raise ExtractorError(msg, expected=True)
450
451 for item in self._extract_items(playlist):
452 kind = item.get('kind')
453 if kind not in ('programme', 'radioProgramme'):
454 continue
455 title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
456 description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
457 description = description_el.text if description_el is not None else None
458
459 def get_programme_id(item):
460 def get_from_attributes(item):
461 for p in('identifier', 'group'):
462 value = item.get(p)
463 if value and re.match(r'^[pb][\da-z]{7}$', value):
464 return value
465 get_from_attributes(item)
466 mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS)
467 if mediator is not None:
468 return get_from_attributes(mediator)
469
470 programme_id = get_programme_id(item)
471 duration = int_or_none(item.get('duration'))
472
473 if programme_id:
474 formats, subtitles = self._download_media_selector(programme_id)
475 else:
476 formats, subtitles = self._process_media_selector(item, playlist_id)
477 programme_id = playlist_id
478
479 return programme_id, title, description, duration, formats, subtitles
480
481 def _real_extract(self, url):
482 group_id = self._match_id(url)
483
484 webpage = self._download_webpage(url, group_id, 'Downloading video page')
485
486 programme_id = None
487 duration = None
488
489 tviplayer = self._search_regex(
490 r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
491 webpage, 'player', default=None)
492
493 if tviplayer:
494 player = self._parse_json(tviplayer, group_id).get('player', {})
495 duration = int_or_none(player.get('duration'))
496 programme_id = player.get('vpid')
497
498 if not programme_id:
499 programme_id = self._search_regex(
500 r'"vpid"\s*:\s*"(%s)"' % self._ID_REGEX, webpage, 'vpid', fatal=False, default=None)
501
502 if programme_id:
503 formats, subtitles = self._download_media_selector(programme_id)
504 title = self._og_search_title(webpage, default=None) or self._html_search_regex(
505 (r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>',
506 r'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage, 'title')
507 description = self._search_regex(
508 (r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
509 r'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'),
510 webpage, 'description', default=None)
511 if not description:
512 description = self._html_search_meta('description', webpage)
513 else:
514 programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
515
516 self._sort_formats(formats)
517
518 return {
519 'id': programme_id,
520 'title': title,
521 'description': description,
522 'thumbnail': self._og_search_thumbnail(webpage, default=None),
523 'duration': duration,
524 'formats': formats,
525 'subtitles': subtitles,
526 }
527
528
529 class BBCIE(BBCCoUkIE):
530 IE_NAME = 'bbc'
531 IE_DESC = 'BBC'
532 _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
533
534 _MEDIASELECTOR_URLS = [
535 # Provides HQ HLS streams but fails with geolocation in some cases when it's
536 # even not geo restricted at all
537 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
538 # Provides more formats, namely direct mp4 links, but fails on some videos with
539 # notukerror for non UK (?) users (e.g.
540 # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
541 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
542 # Provides fewer formats, but works everywhere for everybody (hopefully)
543 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
544 ]
545
546 _TESTS = [{
547 # article with multiple videos embedded with data-playable containing vpids
548 'url': 'http://www.bbc.com/news/world-europe-32668511',
549 'info_dict': {
550 'id': 'world-europe-32668511',
551 'title': 'Russia stages massive WW2 parade despite Western boycott',
552 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
553 },
554 'playlist_count': 2,
555 }, {
556 # article with multiple videos embedded with data-playable (more videos)
557 'url': 'http://www.bbc.com/news/business-28299555',
558 'info_dict': {
559 'id': 'business-28299555',
560 'title': 'Farnborough Airshow: Video highlights',
561 'description': 'BBC reports and video highlights at the Farnborough Airshow.',
562 },
563 'playlist_count': 9,
564 'skip': 'Save time',
565 }, {
566 # article with multiple videos embedded with `new SMP()`
567 # broken
568 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
569 'info_dict': {
570 'id': '3662a707-0af9-3149-963f-47bea720b460',
571 'title': 'BUGGER',
572 },
573 'playlist_count': 18,
574 }, {
575 # single video embedded with data-playable containing vpid
576 'url': 'http://www.bbc.com/news/world-europe-32041533',
577 'info_dict': {
578 'id': 'p02mprgb',
579 'ext': 'mp4',
580 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
581 'description': 'md5:2868290467291b37feda7863f7a83f54',
582 'duration': 47,
583 'timestamp': 1427219242,
584 'upload_date': '20150324',
585 },
586 'params': {
587 # rtmp download
588 'skip_download': True,
589 }
590 }, {
591 # article with single video embedded with data-playable containing XML playlist
592 # with direct video links as progressiveDownloadUrl (for now these are extracted)
593 # and playlist with f4m and m3u8 as streamingUrl
594 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
595 'info_dict': {
596 'id': '150615_telabyad_kentin_cogu',
597 'ext': 'mp4',
598 'title': "YPG: Tel Abyad'ın tamamı kontrolĆ¼mĆ¼zde",
599 'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
600 'timestamp': 1434397334,
601 'upload_date': '20150615',
602 },
603 'params': {
604 'skip_download': True,
605 }
606 }, {
607 # single video embedded with data-playable containing XML playlists (regional section)
608 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
609 'info_dict': {
610 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
611 'ext': 'mp4',
612 'title': 'Honduras militariza sus hospitales por nuevo escĆ”ndalo de corrupciĆ³n',
613 'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
614 'timestamp': 1434713142,
615 'upload_date': '20150619',
616 },
617 'params': {
618 'skip_download': True,
619 }
620 }, {
621 # single video from video playlist embedded with vxp-playlist-data JSON
622 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
623 'info_dict': {
624 'id': 'p02w6qjc',
625 'ext': 'mp4',
626 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
627 'duration': 56,
628 'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
629 },
630 'params': {
631 'skip_download': True,
632 }
633 }, {
634 # single video story with digitalData
635 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
636 'info_dict': {
637 'id': 'p02q6gc4',
638 'ext': 'flv',
639 'title': 'Sri Lankaā€™s spicy secret',
640 'description': 'As a new train line to Jaffna opens up the countryā€™s north, travellers can experience a truly distinct slice of Tamil culture.',
641 'timestamp': 1437674293,
642 'upload_date': '20150723',
643 },
644 'params': {
645 # rtmp download
646 'skip_download': True,
647 }
648 }, {
649 # single video story without digitalData
650 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
651 'info_dict': {
652 'id': 'p018zqqg',
653 'ext': 'mp4',
654 'title': 'Hyundai Santa Fe Sport: Rock star',
655 'description': 'md5:b042a26142c4154a6e472933cf20793d',
656 'timestamp': 1415867444,
657 'upload_date': '20141113',
658 },
659 'params': {
660 # rtmp download
661 'skip_download': True,
662 }
663 }, {
664 # single video embedded with Morph
665 'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
666 'info_dict': {
667 'id': 'p041vhd0',
668 'ext': 'mp4',
669 'title': "Nigeria v Japan - Men's First Round",
670 'description': 'Live coverage of the first round from Group B at the Amazonia Arena.',
671 'duration': 7980,
672 'uploader': 'BBC Sport',
673 'uploader_id': 'bbc_sport',
674 },
675 'params': {
676 # m3u8 download
677 'skip_download': True,
678 },
679 'skip': 'Georestricted to UK',
680 }, {
681 # single video with playlist.sxml URL in playlist param
682 'url': 'http://www.bbc.com/sport/0/football/33653409',
683 'info_dict': {
684 'id': 'p02xycnp',
685 'ext': 'mp4',
686 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
687 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
688 'duration': 140,
689 },
690 'params': {
691 # rtmp download
692 'skip_download': True,
693 }
694 }, {
695 # article with multiple videos embedded with playlist.sxml in playlist param
696 'url': 'http://www.bbc.com/sport/0/football/34475836',
697 'info_dict': {
698 'id': '34475836',
699 'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
700 'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
701 },
702 'playlist_count': 3,
703 }, {
704 # school report article with single video
705 'url': 'http://www.bbc.co.uk/schoolreport/35744779',
706 'info_dict': {
707 'id': '35744779',
708 'title': 'School which breaks down barriers in Jerusalem',
709 },
710 'playlist_count': 1,
711 }, {
712 # single video with playlist URL from weather section
713 'url': 'http://www.bbc.com/weather/features/33601775',
714 'only_matching': True,
715 }, {
716 # custom redirection to www.bbc.com
717 'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
718 'only_matching': True,
719 }, {
720 # single video article embedded with data-media-vpid
721 'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
722 'only_matching': True,
723 }]
724
725 @classmethod
726 def suitable(cls, url):
727 EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE)
728 return (False if any(ie.suitable(url) for ie in EXCLUDE_IE)
729 else super(BBCIE, cls).suitable(url))
730
731 def _extract_from_media_meta(self, media_meta, video_id):
732 # Direct links to media in media metadata (e.g.
733 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
734 # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
735 source_files = media_meta.get('sourceFiles')
736 if source_files:
737 return [{
738 'url': f['url'],
739 'format_id': format_id,
740 'ext': f.get('encoding'),
741 'tbr': float_or_none(f.get('bitrate'), 1000),
742 'filesize': int_or_none(f.get('filesize')),
743 } for format_id, f in source_files.items() if f.get('url')], []
744
745 programme_id = media_meta.get('externalId')
746 if programme_id:
747 return self._download_media_selector(programme_id)
748
749 # Process playlist.sxml as legacy playlist
750 href = media_meta.get('href')
751 if href:
752 playlist = self._download_legacy_playlist_url(href)
753 _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
754 return formats, subtitles
755
756 return [], []
757
758 def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
759 programme_id, title, description, duration, formats, subtitles = \
760 self._process_legacy_playlist_url(url, playlist_id)
761 self._sort_formats(formats)
762 return {
763 'id': programme_id,
764 'title': title,
765 'description': description,
766 'duration': duration,
767 'timestamp': timestamp,
768 'formats': formats,
769 'subtitles': subtitles,
770 }
771
772 def _real_extract(self, url):
773 playlist_id = self._match_id(url)
774
775 webpage = self._download_webpage(url, playlist_id)
776
777 json_ld_info = self._search_json_ld(webpage, playlist_id, default={})
778 timestamp = json_ld_info.get('timestamp')
779
780 playlist_title = json_ld_info.get('title')
781 if not playlist_title:
782 playlist_title = self._og_search_title(
783 webpage, default=None) or self._html_search_regex(
784 r'<title>(.+?)</title>', webpage, 'playlist title', default=None)
785 if playlist_title:
786 playlist_title = re.sub(r'(.+)\s*-\s*BBC.*?$', r'\1', playlist_title).strip()
787
788 playlist_description = json_ld_info.get(
789 'description') or self._og_search_description(webpage, default=None)
790
791 if not timestamp:
792 timestamp = parse_iso8601(self._search_regex(
793 [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
794 r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
795 r'"datePublished":\s*"([^"]+)'],
796 webpage, 'date', default=None))
797
798 entries = []
799
800 # article with multiple videos embedded with playlist.sxml (e.g.
801 # http://www.bbc.com/sport/0/football/34475836)
802 playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
803 playlists.extend(re.findall(r'data-media-id="([^"]+/playlist\.sxml)"', webpage))
804 if playlists:
805 entries = [
806 self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
807 for playlist_url in playlists]
808
809 # news article with multiple videos embedded with data-playable
810 data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
811 if data_playables:
812 for _, data_playable_json in data_playables:
813 data_playable = self._parse_json(
814 unescapeHTML(data_playable_json), playlist_id, fatal=False)
815 if not data_playable:
816 continue
817 settings = data_playable.get('settings', {})
818 if settings:
819 # data-playable with video vpid in settings.playlistObject.items (e.g.
820 # http://www.bbc.com/news/world-us-canada-34473351)
821 playlist_object = settings.get('playlistObject', {})
822 if playlist_object:
823 items = playlist_object.get('items')
824 if items and isinstance(items, list):
825 title = playlist_object['title']
826 description = playlist_object.get('summary')
827 duration = int_or_none(items[0].get('duration'))
828 programme_id = items[0].get('vpid')
829 formats, subtitles = self._download_media_selector(programme_id)
830 self._sort_formats(formats)
831 entries.append({
832 'id': programme_id,
833 'title': title,
834 'description': description,
835 'timestamp': timestamp,
836 'duration': duration,
837 'formats': formats,
838 'subtitles': subtitles,
839 })
840 else:
841 # data-playable without vpid but with a playlist.sxml URLs
842 # in otherSettings.playlist (e.g.
843 # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
844 playlist = data_playable.get('otherSettings', {}).get('playlist', {})
845 if playlist:
846 entry = None
847 for key in ('streaming', 'progressiveDownload'):
848 playlist_url = playlist.get('%sUrl' % key)
849 if not playlist_url:
850 continue
851 try:
852 info = self._extract_from_playlist_sxml(
853 playlist_url, playlist_id, timestamp)
854 if not entry:
855 entry = info
856 else:
857 entry['title'] = info['title']
858 entry['formats'].extend(info['formats'])
859 except Exception as e:
860 # Some playlist URL may fail with 500, at the same time
861 # the other one may work fine (e.g.
862 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
863 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
864 continue
865 raise
866 if entry:
867 self._sort_formats(entry['formats'])
868 entries.append(entry)
869
870 if entries:
871 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
872
873 # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
874 programme_id = self._search_regex(
875 [r'data-(?:video-player|media)-vpid="(%s)"' % self._ID_REGEX,
876 r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
877 r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
878 webpage, 'vpid', default=None)
879
880 if programme_id:
881 formats, subtitles = self._download_media_selector(programme_id)
882 self._sort_formats(formats)
883 # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
884 digital_data = self._parse_json(
885 self._search_regex(
886 r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
887 programme_id, fatal=False)
888 page_info = digital_data.get('page', {}).get('pageInfo', {})
889 title = page_info.get('pageName') or self._og_search_title(webpage)
890 description = page_info.get('description') or self._og_search_description(webpage)
891 timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
892 return {
893 'id': programme_id,
894 'title': title,
895 'description': description,
896 'timestamp': timestamp,
897 'formats': formats,
898 'subtitles': subtitles,
899 }
900
901 # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
902 # There are several setPayload calls may be present but the video
903 # seems to be always related to the first one
904 morph_payload = self._parse_json(
905 self._search_regex(
906 r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
907 webpage, 'morph payload', default='{}'),
908 playlist_id, fatal=False)
909 if morph_payload:
910 components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
911 for component in components:
912 if not isinstance(component, dict):
913 continue
914 lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
915 if not lead_media:
916 continue
917 identifiers = lead_media.get('identifiers')
918 if not identifiers or not isinstance(identifiers, dict):
919 continue
920 programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
921 if not programme_id:
922 continue
923 title = lead_media.get('title') or self._og_search_title(webpage)
924 formats, subtitles = self._download_media_selector(programme_id)
925 self._sort_formats(formats)
926 description = lead_media.get('summary')
927 uploader = lead_media.get('masterBrand')
928 uploader_id = lead_media.get('mid')
929 duration = None
930 duration_d = lead_media.get('duration')
931 if isinstance(duration_d, dict):
932 duration = parse_duration(dict_get(
933 duration_d, ('rawDuration', 'formattedDuration', 'spokenDuration')))
934 return {
935 'id': programme_id,
936 'title': title,
937 'description': description,
938 'duration': duration,
939 'uploader': uploader,
940 'uploader_id': uploader_id,
941 'formats': formats,
942 'subtitles': subtitles,
943 }
944
945 def extract_all(pattern):
946 return list(filter(None, map(
947 lambda s: self._parse_json(s, playlist_id, fatal=False),
948 re.findall(pattern, webpage))))
949
950 # Multiple video article (e.g.
951 # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
952 EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX
953 entries = []
954 for match in extract_all(r'new\s+SMP\(({.+?})\)'):
955 embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
956 if embed_url and re.match(EMBED_URL, embed_url):
957 entries.append(embed_url)
958 entries.extend(re.findall(
959 r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
960 if entries:
961 return self.playlist_result(
962 [self.url_result(entry_, 'BBCCoUk') for entry_ in entries],
963 playlist_id, playlist_title, playlist_description)
964
965 # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
966 medias = extract_all(r"data-media-meta='({[^']+})'")
967
968 if not medias:
969 # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
970 media_asset = self._search_regex(
971 r'mediaAssetPage\.init\(\s*({.+?}), "/',
972 webpage, 'media asset', default=None)
973 if media_asset:
974 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
975 medias = []
976 for video in media_asset_page.get('videos', {}).values():
977 medias.extend(video.values())
978
979 if not medias:
980 # Multiple video playlist with single `now playing` entry (e.g.
981 # http://www.bbc.com/news/video_and_audio/must_see/33767813)
982 vxp_playlist = self._parse_json(
983 self._search_regex(
984 r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>',
985 webpage, 'playlist data'),
986 playlist_id)
987 playlist_medias = []
988 for item in vxp_playlist:
989 media = item.get('media')
990 if not media:
991 continue
992 playlist_medias.append(media)
993 # Download single video if found media with asset id matching the video id from URL
994 if item.get('advert', {}).get('assetId') == playlist_id:
995 medias = [media]
996 break
997 # Fallback to the whole playlist
998 if not medias:
999 medias = playlist_medias
1000
1001 entries = []
1002 for num, media_meta in enumerate(medias, start=1):
1003 formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
1004 if not formats:
1005 continue
1006 self._sort_formats(formats)
1007
1008 video_id = media_meta.get('externalId')
1009 if not video_id:
1010 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
1011
1012 title = media_meta.get('caption')
1013 if not title:
1014 title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
1015
1016 duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
1017
1018 images = []
1019 for image in media_meta.get('images', {}).values():
1020 images.extend(image.values())
1021 if 'image' in media_meta:
1022 images.append(media_meta['image'])
1023
1024 thumbnails = [{
1025 'url': image.get('href'),
1026 'width': int_or_none(image.get('width')),
1027 'height': int_or_none(image.get('height')),
1028 } for image in images]
1029
1030 entries.append({
1031 'id': video_id,
1032 'title': title,
1033 'thumbnails': thumbnails,
1034 'duration': duration,
1035 'timestamp': timestamp,
1036 'formats': formats,
1037 'subtitles': subtitles,
1038 })
1039
1040 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
1041
1042
1043 class BBCCoUkArticleIE(InfoExtractor):
1044 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
1045 IE_NAME = 'bbc.co.uk:article'
1046 IE_DESC = 'BBC articles'
1047
1048 _TEST = {
1049 'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer',
1050 'info_dict': {
1051 'id': '3jNQLTMrPlYGTBn0WV6M2MS',
1052 'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four',
1053 'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.',
1054 },
1055 'playlist_count': 4,
1056 'add_ie': ['BBCCoUk'],
1057 }
1058
1059 def _real_extract(self, url):
1060 playlist_id = self._match_id(url)
1061
1062 webpage = self._download_webpage(url, playlist_id)
1063
1064 title = self._og_search_title(webpage)
1065 description = self._og_search_description(webpage).strip()
1066
1067 entries = [self.url_result(programme_url) for programme_url in re.findall(
1068 r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
1069
1070 return self.playlist_result(entries, playlist_id, title, description)
1071
1072
1073 class BBCCoUkPlaylistBaseIE(InfoExtractor):
1074 def _entries(self, webpage, url, playlist_id):
1075 single_page = 'page' in compat_urlparse.parse_qs(
1076 compat_urlparse.urlparse(url).query)
1077 for page_num in itertools.count(2):
1078 for video_id in re.findall(
1079 self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage):
1080 yield self.url_result(
1081 self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
1082 if single_page:
1083 return
1084 next_page = self._search_regex(
1085 r'<li[^>]+class=(["\'])pagination_+next\1[^>]*><a[^>]+href=(["\'])(?P<url>(?:(?!\2).)+)\2',
1086 webpage, 'next page url', default=None, group='url')
1087 if not next_page:
1088 break
1089 webpage = self._download_webpage(
1090 compat_urlparse.urljoin(url, next_page), playlist_id,
1091 'Downloading page %d' % page_num, page_num)
1092
1093 def _real_extract(self, url):
1094 playlist_id = self._match_id(url)
1095
1096 webpage = self._download_webpage(url, playlist_id)
1097
1098 title, description = self._extract_title_and_description(webpage)
1099
1100 return self.playlist_result(
1101 self._entries(webpage, url, playlist_id),
1102 playlist_id, title, description)
1103
1104
1105 class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
1106 IE_NAME = 'bbc.co.uk:iplayer:playlist'
1107 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
1108 _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
1109 _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
1110 _TESTS = [{
1111 'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v',
1112 'info_dict': {
1113 'id': 'b05rcz9v',
1114 'title': 'The Disappearance',
1115 'description': 'French thriller serial about a missing teenager.',
1116 },
1117 'playlist_mincount': 6,
1118 'skip': 'This programme is not currently available on BBC iPlayer',
1119 }, {
1120 # Available for over a year unlike 30 days for most other programmes
1121 'url': 'http://www.bbc.co.uk/iplayer/group/p02tcc32',
1122 'info_dict': {
1123 'id': 'p02tcc32',
1124 'title': 'Bohemian Icons',
1125 'description': 'md5:683e901041b2fe9ba596f2ab04c4dbe7',
1126 },
1127 'playlist_mincount': 10,
1128 }]
1129
1130 def _extract_title_and_description(self, webpage):
1131 title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
1132 description = self._search_regex(
1133 r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
1134 webpage, 'description', fatal=False, group='value')
1135 return title, description
1136
1137
1138 class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
1139 IE_NAME = 'bbc.co.uk:playlist'
1140 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
1141 _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
1142 _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
1143 _TESTS = [{
1144 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1145 'info_dict': {
1146 'id': 'b05rcz9v',
1147 'title': 'The Disappearance - Clips - BBC Four',
1148 'description': 'French thriller serial about a missing teenager.',
1149 },
1150 'playlist_mincount': 7,
1151 }, {
1152 # multipage playlist, explicit page
1153 'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips?page=1',
1154 'info_dict': {
1155 'id': 'b00mfl7n',
1156 'title': 'Frozen Planet - Clips - BBC One',
1157 'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1158 },
1159 'playlist_mincount': 24,
1160 }, {
1161 # multipage playlist, all pages
1162 'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips',
1163 'info_dict': {
1164 'id': 'b00mfl7n',
1165 'title': 'Frozen Planet - Clips - BBC One',
1166 'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1167 },
1168 'playlist_mincount': 142,
1169 }, {
1170 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/broadcasts/2016/06',
1171 'only_matching': True,
1172 }, {
1173 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1174 'only_matching': True,
1175 }, {
1176 'url': 'http://www.bbc.co.uk/programmes/b055jkys/episodes/player',
1177 'only_matching': True,
1178 }]
1179
1180 def _extract_title_and_description(self, webpage):
1181 title = self._og_search_title(webpage, fatal=False)
1182 description = self._og_search_description(webpage)
1183 return title, description