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 class MediaSelectionError(Exception):
229 def __init__(self, id):
230 self.id = id
231
232 def _extract_asx_playlist(self, connection, programme_id):
233 asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
234 return [ref.get('href') for ref in asx.findall('./Entry/ref')]
235
236 def _extract_items(self, playlist):
237 return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
238
239 def _findall_ns(self, element, xpath):
240 elements = []
241 for ns in self._NAMESPACES:
242 elements.extend(element.findall(xpath % ns))
243 return elements
244
245 def _extract_medias(self, media_selection):
246 error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS)
247 if error is None:
248 media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS)
249 if error is not None:
250 raise BBCCoUkIE.MediaSelectionError(error.get('id'))
251 return self._findall_ns(media_selection, './{%s}media')
252
253 def _extract_connections(self, media):
254 return self._findall_ns(media, './{%s}connection')
255
256 def _get_subtitles(self, media, programme_id):
257 subtitles = {}
258 for connection in self._extract_connections(media):
259 captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
260 lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
261 subtitles[lang] = [
262 {
263 'url': connection.get('href'),
264 'ext': 'ttml',
265 },
266 ]
267 return subtitles
268
269 def _raise_extractor_error(self, media_selection_error):
270 raise ExtractorError(
271 '%s returned error: %s' % (self.IE_NAME, media_selection_error.id),
272 expected=True)
273
274 def _download_media_selector(self, programme_id):
275 last_exception = None
276 for mediaselector_url in self._MEDIASELECTOR_URLS:
277 try:
278 return self._download_media_selector_url(
279 mediaselector_url % programme_id, programme_id)
280 except BBCCoUkIE.MediaSelectionError as e:
281 if e.id in ('notukerror', 'geolocation', 'selectionunavailable'):
282 last_exception = e
283 continue
284 self._raise_extractor_error(e)
285 self._raise_extractor_error(last_exception)
286
287 def _download_media_selector_url(self, url, programme_id=None):
288 try:
289 media_selection = self._download_xml(
290 url, programme_id, 'Downloading media selection XML')
291 except ExtractorError as ee:
292 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code in (403, 404):
293 media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8'))
294 else:
295 raise
296 return self._process_media_selector(media_selection, programme_id)
297
298 def _process_media_selector(self, media_selection, programme_id):
299 formats = []
300 subtitles = None
301 urls = []
302
303 for media in self._extract_medias(media_selection):
304 kind = media.get('kind')
305 if kind in ('video', 'audio'):
306 bitrate = int_or_none(media.get('bitrate'))
307 encoding = media.get('encoding')
308 service = media.get('service')
309 width = int_or_none(media.get('width'))
310 height = int_or_none(media.get('height'))
311 file_size = int_or_none(media.get('media_file_size'))
312 for connection in self._extract_connections(media):
313 href = connection.get('href')
314 if href in urls:
315 continue
316 if href:
317 urls.append(href)
318 conn_kind = connection.get('kind')
319 protocol = connection.get('protocol')
320 supplier = connection.get('supplier')
321 transfer_format = connection.get('transferFormat')
322 format_id = supplier or conn_kind or protocol
323 if service:
324 format_id = '%s_%s' % (service, format_id)
325 # ASX playlist
326 if supplier == 'asx':
327 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
328 formats.append({
329 'url': ref,
330 'format_id': 'ref%s_%s' % (i, format_id),
331 })
332 elif transfer_format == 'dash':
333 formats.extend(self._extract_mpd_formats(
334 href, programme_id, mpd_id=format_id, fatal=False))
335 elif transfer_format == 'hls':
336 formats.extend(self._extract_m3u8_formats(
337 href, programme_id, ext='mp4', entry_protocol='m3u8_native',
338 m3u8_id=format_id, fatal=False))
339 elif transfer_format == 'hds':
340 formats.extend(self._extract_f4m_formats(
341 href, programme_id, f4m_id=format_id, fatal=False))
342 else:
343 if not service and not supplier and bitrate:
344 format_id += '-%d' % bitrate
345 fmt = {
346 'format_id': format_id,
347 'filesize': file_size,
348 }
349 if kind == 'video':
350 fmt.update({
351 'width': width,
352 'height': height,
353 'vbr': bitrate,
354 'vcodec': encoding,
355 })
356 else:
357 fmt.update({
358 'abr': bitrate,
359 'acodec': encoding,
360 'vcodec': 'none',
361 })
362 if protocol == 'http':
363 # Direct link
364 fmt.update({
365 'url': href,
366 })
367 elif protocol == 'rtmp':
368 application = connection.get('application', 'ondemand')
369 auth_string = connection.get('authString')
370 identifier = connection.get('identifier')
371 server = connection.get('server')
372 fmt.update({
373 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
374 'play_path': identifier,
375 'app': '%s?%s' % (application, auth_string),
376 'page_url': 'http://www.bbc.co.uk',
377 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
378 'rtmp_live': False,
379 'ext': 'flv',
380 })
381 formats.append(fmt)
382 elif kind == 'captions':
383 subtitles = self.extract_subtitles(media, programme_id)
384 return formats, subtitles
385
386 def _download_playlist(self, playlist_id):
387 try:
388 playlist = self._download_json(
389 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
390 playlist_id, 'Downloading playlist JSON')
391
392 version = playlist.get('defaultAvailableVersion')
393 if version:
394 smp_config = version['smpConfig']
395 title = smp_config['title']
396 description = smp_config['summary']
397 for item in smp_config['items']:
398 kind = item['kind']
399 if kind != 'programme' and kind != 'radioProgramme':
400 continue
401 programme_id = item.get('vpid')
402 duration = int_or_none(item.get('duration'))
403 formats, subtitles = self._download_media_selector(programme_id)
404 return programme_id, title, description, duration, formats, subtitles
405 except ExtractorError as ee:
406 if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
407 raise
408
409 # fallback to legacy playlist
410 return self._process_legacy_playlist(playlist_id)
411
412 def _process_legacy_playlist_url(self, url, display_id):
413 playlist = self._download_legacy_playlist_url(url, display_id)
414 return self._extract_from_legacy_playlist(playlist, display_id)
415
416 def _process_legacy_playlist(self, playlist_id):
417 return self._process_legacy_playlist_url(
418 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
419
420 def _download_legacy_playlist_url(self, url, playlist_id=None):
421 return self._download_xml(
422 url, playlist_id, 'Downloading legacy playlist XML')
423
424 def _extract_from_legacy_playlist(self, playlist, playlist_id):
425 no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS)
426 if no_items is not None:
427 reason = no_items.get('reason')
428 if reason == 'preAvailability':
429 msg = 'Episode %s is not yet available' % playlist_id
430 elif reason == 'postAvailability':
431 msg = 'Episode %s is no longer available' % playlist_id
432 elif reason == 'noMedia':
433 msg = 'Episode %s is not currently available' % playlist_id
434 else:
435 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
436 raise ExtractorError(msg, expected=True)
437
438 for item in self._extract_items(playlist):
439 kind = item.get('kind')
440 if kind != 'programme' and kind != 'radioProgramme':
441 continue
442 title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
443 description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
444 description = description_el.text if description_el is not None else None
445
446 def get_programme_id(item):
447 def get_from_attributes(item):
448 for p in('identifier', 'group'):
449 value = item.get(p)
450 if value and re.match(r'^[pb][\da-z]{7}$', value):
451 return value
452 get_from_attributes(item)
453 mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS)
454 if mediator is not None:
455 return get_from_attributes(mediator)
456
457 programme_id = get_programme_id(item)
458 duration = int_or_none(item.get('duration'))
459
460 if programme_id:
461 formats, subtitles = self._download_media_selector(programme_id)
462 else:
463 formats, subtitles = self._process_media_selector(item, playlist_id)
464 programme_id = playlist_id
465
466 return programme_id, title, description, duration, formats, subtitles
467
468 def _real_extract(self, url):
469 group_id = self._match_id(url)
470
471 webpage = self._download_webpage(url, group_id, 'Downloading video page')
472
473 programme_id = None
474 duration = None
475
476 tviplayer = self._search_regex(
477 r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
478 webpage, 'player', default=None)
479
480 if tviplayer:
481 player = self._parse_json(tviplayer, group_id).get('player', {})
482 duration = int_or_none(player.get('duration'))
483 programme_id = player.get('vpid')
484
485 if not programme_id:
486 programme_id = self._search_regex(
487 r'"vpid"\s*:\s*"(%s)"' % self._ID_REGEX, webpage, 'vpid', fatal=False, default=None)
488
489 if programme_id:
490 formats, subtitles = self._download_media_selector(programme_id)
491 title = self._og_search_title(webpage, default=None) or self._html_search_regex(
492 (r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>',
493 r'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage, 'title')
494 description = self._search_regex(
495 (r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
496 r'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'),
497 webpage, 'description', default=None)
498 if not description:
499 description = self._html_search_meta('description', webpage)
500 else:
501 programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
502
503 self._sort_formats(formats)
504
505 return {
506 'id': programme_id,
507 'title': title,
508 'description': description,
509 'thumbnail': self._og_search_thumbnail(webpage, default=None),
510 'duration': duration,
511 'formats': formats,
512 'subtitles': subtitles,
513 }
514
515
516 class BBCIE(BBCCoUkIE):
517 IE_NAME = 'bbc'
518 IE_DESC = 'BBC'
519 _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
520
521 _MEDIASELECTOR_URLS = [
522 # Provides HQ HLS streams but fails with geolocation in some cases when it's
523 # even not geo restricted at all
524 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
525 # Provides more formats, namely direct mp4 links, but fails on some videos with
526 # notukerror for non UK (?) users (e.g.
527 # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
528 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
529 # Provides fewer formats, but works everywhere for everybody (hopefully)
530 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
531 ]
532
533 _TESTS = [{
534 # article with multiple videos embedded with data-playable containing vpids
535 'url': 'http://www.bbc.com/news/world-europe-32668511',
536 'info_dict': {
537 'id': 'world-europe-32668511',
538 'title': 'Russia stages massive WW2 parade despite Western boycott',
539 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
540 },
541 'playlist_count': 2,
542 }, {
543 # article with multiple videos embedded with data-playable (more videos)
544 'url': 'http://www.bbc.com/news/business-28299555',
545 'info_dict': {
546 'id': 'business-28299555',
547 'title': 'Farnborough Airshow: Video highlights',
548 'description': 'BBC reports and video highlights at the Farnborough Airshow.',
549 },
550 'playlist_count': 9,
551 'skip': 'Save time',
552 }, {
553 # article with multiple videos embedded with `new SMP()`
554 # broken
555 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
556 'info_dict': {
557 'id': '3662a707-0af9-3149-963f-47bea720b460',
558 'title': 'BUGGER',
559 },
560 'playlist_count': 18,
561 }, {
562 # single video embedded with data-playable containing vpid
563 'url': 'http://www.bbc.com/news/world-europe-32041533',
564 'info_dict': {
565 'id': 'p02mprgb',
566 'ext': 'mp4',
567 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
568 'description': 'md5:2868290467291b37feda7863f7a83f54',
569 'duration': 47,
570 'timestamp': 1427219242,
571 'upload_date': '20150324',
572 },
573 'params': {
574 # rtmp download
575 'skip_download': True,
576 }
577 }, {
578 # article with single video embedded with data-playable containing XML playlist
579 # with direct video links as progressiveDownloadUrl (for now these are extracted)
580 # and playlist with f4m and m3u8 as streamingUrl
581 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
582 'info_dict': {
583 'id': '150615_telabyad_kentin_cogu',
584 'ext': 'mp4',
585 'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
586 'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
587 'timestamp': 1434397334,
588 'upload_date': '20150615',
589 },
590 'params': {
591 'skip_download': True,
592 }
593 }, {
594 # single video embedded with data-playable containing XML playlists (regional section)
595 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
596 'info_dict': {
597 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
598 'ext': 'mp4',
599 'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
600 'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
601 'timestamp': 1434713142,
602 'upload_date': '20150619',
603 },
604 'params': {
605 'skip_download': True,
606 }
607 }, {
608 # single video from video playlist embedded with vxp-playlist-data JSON
609 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
610 'info_dict': {
611 'id': 'p02w6qjc',
612 'ext': 'mp4',
613 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
614 'duration': 56,
615 'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
616 },
617 'params': {
618 'skip_download': True,
619 }
620 }, {
621 # single video story with digitalData
622 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
623 'info_dict': {
624 'id': 'p02q6gc4',
625 'ext': 'flv',
626 'title': 'Sri Lanka’s spicy secret',
627 'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
628 'timestamp': 1437674293,
629 'upload_date': '20150723',
630 },
631 'params': {
632 # rtmp download
633 'skip_download': True,
634 }
635 }, {
636 # single video story without digitalData
637 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
638 'info_dict': {
639 'id': 'p018zqqg',
640 'ext': 'mp4',
641 'title': 'Hyundai Santa Fe Sport: Rock star',
642 'description': 'md5:b042a26142c4154a6e472933cf20793d',
643 'timestamp': 1415867444,
644 'upload_date': '20141113',
645 },
646 'params': {
647 # rtmp download
648 'skip_download': True,
649 }
650 }, {
651 # single video embedded with Morph
652 'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
653 'info_dict': {
654 'id': 'p041vhd0',
655 'ext': 'mp4',
656 'title': "Nigeria v Japan - Men's First Round",
657 'description': 'Live coverage of the first round from Group B at the Amazonia Arena.',
658 'duration': 7980,
659 'uploader': 'BBC Sport',
660 'uploader_id': 'bbc_sport',
661 },
662 'params': {
663 # m3u8 download
664 'skip_download': True,
665 },
666 'skip': 'Georestricted to UK',
667 }, {
668 # single video with playlist.sxml URL in playlist param
669 'url': 'http://www.bbc.com/sport/0/football/33653409',
670 'info_dict': {
671 'id': 'p02xycnp',
672 'ext': 'mp4',
673 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
674 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
675 'duration': 140,
676 },
677 'params': {
678 # rtmp download
679 'skip_download': True,
680 }
681 }, {
682 # article with multiple videos embedded with playlist.sxml in playlist param
683 'url': 'http://www.bbc.com/sport/0/football/34475836',
684 'info_dict': {
685 'id': '34475836',
686 'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
687 'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
688 },
689 'playlist_count': 3,
690 }, {
691 # school report article with single video
692 'url': 'http://www.bbc.co.uk/schoolreport/35744779',
693 'info_dict': {
694 'id': '35744779',
695 'title': 'School which breaks down barriers in Jerusalem',
696 },
697 'playlist_count': 1,
698 }, {
699 # single video with playlist URL from weather section
700 'url': 'http://www.bbc.com/weather/features/33601775',
701 'only_matching': True,
702 }, {
703 # custom redirection to www.bbc.com
704 'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
705 'only_matching': True,
706 }, {
707 # single video article embedded with data-media-vpid
708 'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
709 'only_matching': True,
710 }]
711
712 @classmethod
713 def suitable(cls, url):
714 EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE)
715 return (False if any(ie.suitable(url) for ie in EXCLUDE_IE)
716 else super(BBCIE, cls).suitable(url))
717
718 def _extract_from_media_meta(self, media_meta, video_id):
719 # Direct links to media in media metadata (e.g.
720 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
721 # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
722 source_files = media_meta.get('sourceFiles')
723 if source_files:
724 return [{
725 'url': f['url'],
726 'format_id': format_id,
727 'ext': f.get('encoding'),
728 'tbr': float_or_none(f.get('bitrate'), 1000),
729 'filesize': int_or_none(f.get('filesize')),
730 } for format_id, f in source_files.items() if f.get('url')], []
731
732 programme_id = media_meta.get('externalId')
733 if programme_id:
734 return self._download_media_selector(programme_id)
735
736 # Process playlist.sxml as legacy playlist
737 href = media_meta.get('href')
738 if href:
739 playlist = self._download_legacy_playlist_url(href)
740 _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
741 return formats, subtitles
742
743 return [], []
744
745 def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
746 programme_id, title, description, duration, formats, subtitles = \
747 self._process_legacy_playlist_url(url, playlist_id)
748 self._sort_formats(formats)
749 return {
750 'id': programme_id,
751 'title': title,
752 'description': description,
753 'duration': duration,
754 'timestamp': timestamp,
755 'formats': formats,
756 'subtitles': subtitles,
757 }
758
759 def _real_extract(self, url):
760 playlist_id = self._match_id(url)
761
762 webpage = self._download_webpage(url, playlist_id)
763
764 json_ld_info = self._search_json_ld(webpage, playlist_id, default={})
765 timestamp = json_ld_info.get('timestamp')
766
767 playlist_title = json_ld_info.get('title')
768 if not playlist_title:
769 playlist_title = self._og_search_title(
770 webpage, default=None) or self._html_search_regex(
771 r'<title>(.+?)</title>', webpage, 'playlist title', default=None)
772 if playlist_title:
773 playlist_title = re.sub(r'(.+)\s*-\s*BBC.*?$', r'\1', playlist_title).strip()
774
775 playlist_description = json_ld_info.get(
776 'description') or self._og_search_description(webpage, default=None)
777
778 if not timestamp:
779 timestamp = parse_iso8601(self._search_regex(
780 [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
781 r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
782 r'"datePublished":\s*"([^"]+)'],
783 webpage, 'date', default=None))
784
785 entries = []
786
787 # article with multiple videos embedded with playlist.sxml (e.g.
788 # http://www.bbc.com/sport/0/football/34475836)
789 playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
790 playlists.extend(re.findall(r'data-media-id="([^"]+/playlist\.sxml)"', webpage))
791 if playlists:
792 entries = [
793 self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
794 for playlist_url in playlists]
795
796 # news article with multiple videos embedded with data-playable
797 data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
798 if data_playables:
799 for _, data_playable_json in data_playables:
800 data_playable = self._parse_json(
801 unescapeHTML(data_playable_json), playlist_id, fatal=False)
802 if not data_playable:
803 continue
804 settings = data_playable.get('settings', {})
805 if settings:
806 # data-playable with video vpid in settings.playlistObject.items (e.g.
807 # http://www.bbc.com/news/world-us-canada-34473351)
808 playlist_object = settings.get('playlistObject', {})
809 if playlist_object:
810 items = playlist_object.get('items')
811 if items and isinstance(items, list):
812 title = playlist_object['title']
813 description = playlist_object.get('summary')
814 duration = int_or_none(items[0].get('duration'))
815 programme_id = items[0].get('vpid')
816 formats, subtitles = self._download_media_selector(programme_id)
817 self._sort_formats(formats)
818 entries.append({
819 'id': programme_id,
820 'title': title,
821 'description': description,
822 'timestamp': timestamp,
823 'duration': duration,
824 'formats': formats,
825 'subtitles': subtitles,
826 })
827 else:
828 # data-playable without vpid but with a playlist.sxml URLs
829 # in otherSettings.playlist (e.g.
830 # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
831 playlist = data_playable.get('otherSettings', {}).get('playlist', {})
832 if playlist:
833 entry = None
834 for key in ('streaming', 'progressiveDownload'):
835 playlist_url = playlist.get('%sUrl' % key)
836 if not playlist_url:
837 continue
838 try:
839 info = self._extract_from_playlist_sxml(
840 playlist_url, playlist_id, timestamp)
841 if not entry:
842 entry = info
843 else:
844 entry['title'] = info['title']
845 entry['formats'].extend(info['formats'])
846 except Exception as e:
847 # Some playlist URL may fail with 500, at the same time
848 # the other one may work fine (e.g.
849 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
850 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
851 continue
852 raise
853 if entry:
854 self._sort_formats(entry['formats'])
855 entries.append(entry)
856
857 if entries:
858 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
859
860 # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
861 programme_id = self._search_regex(
862 [r'data-(?:video-player|media)-vpid="(%s)"' % self._ID_REGEX,
863 r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
864 r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
865 webpage, 'vpid', default=None)
866
867 if programme_id:
868 formats, subtitles = self._download_media_selector(programme_id)
869 self._sort_formats(formats)
870 # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
871 digital_data = self._parse_json(
872 self._search_regex(
873 r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
874 programme_id, fatal=False)
875 page_info = digital_data.get('page', {}).get('pageInfo', {})
876 title = page_info.get('pageName') or self._og_search_title(webpage)
877 description = page_info.get('description') or self._og_search_description(webpage)
878 timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
879 return {
880 'id': programme_id,
881 'title': title,
882 'description': description,
883 'timestamp': timestamp,
884 'formats': formats,
885 'subtitles': subtitles,
886 }
887
888 # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
889 # There are several setPayload calls may be present but the video
890 # seems to be always related to the first one
891 morph_payload = self._parse_json(
892 self._search_regex(
893 r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
894 webpage, 'morph payload', default='{}'),
895 playlist_id, fatal=False)
896 if morph_payload:
897 components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
898 for component in components:
899 if not isinstance(component, dict):
900 continue
901 lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
902 if not lead_media:
903 continue
904 identifiers = lead_media.get('identifiers')
905 if not identifiers or not isinstance(identifiers, dict):
906 continue
907 programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
908 if not programme_id:
909 continue
910 title = lead_media.get('title') or self._og_search_title(webpage)
911 formats, subtitles = self._download_media_selector(programme_id)
912 self._sort_formats(formats)
913 description = lead_media.get('summary')
914 uploader = lead_media.get('masterBrand')
915 uploader_id = lead_media.get('mid')
916 duration = None
917 duration_d = lead_media.get('duration')
918 if isinstance(duration_d, dict):
919 duration = parse_duration(dict_get(
920 duration_d, ('rawDuration', 'formattedDuration', 'spokenDuration')))
921 return {
922 'id': programme_id,
923 'title': title,
924 'description': description,
925 'duration': duration,
926 'uploader': uploader,
927 'uploader_id': uploader_id,
928 'formats': formats,
929 'subtitles': subtitles,
930 }
931
932 def extract_all(pattern):
933 return list(filter(None, map(
934 lambda s: self._parse_json(s, playlist_id, fatal=False),
935 re.findall(pattern, webpage))))
936
937 # Multiple video article (e.g.
938 # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
939 EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX
940 entries = []
941 for match in extract_all(r'new\s+SMP\(({.+?})\)'):
942 embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
943 if embed_url and re.match(EMBED_URL, embed_url):
944 entries.append(embed_url)
945 entries.extend(re.findall(
946 r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
947 if entries:
948 return self.playlist_result(
949 [self.url_result(entry_, 'BBCCoUk') for entry_ in entries],
950 playlist_id, playlist_title, playlist_description)
951
952 # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
953 medias = extract_all(r"data-media-meta='({[^']+})'")
954
955 if not medias:
956 # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
957 media_asset = self._search_regex(
958 r'mediaAssetPage\.init\(\s*({.+?}), "/',
959 webpage, 'media asset', default=None)
960 if media_asset:
961 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
962 medias = []
963 for video in media_asset_page.get('videos', {}).values():
964 medias.extend(video.values())
965
966 if not medias:
967 # Multiple video playlist with single `now playing` entry (e.g.
968 # http://www.bbc.com/news/video_and_audio/must_see/33767813)
969 vxp_playlist = self._parse_json(
970 self._search_regex(
971 r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>',
972 webpage, 'playlist data'),
973 playlist_id)
974 playlist_medias = []
975 for item in vxp_playlist:
976 media = item.get('media')
977 if not media:
978 continue
979 playlist_medias.append(media)
980 # Download single video if found media with asset id matching the video id from URL
981 if item.get('advert', {}).get('assetId') == playlist_id:
982 medias = [media]
983 break
984 # Fallback to the whole playlist
985 if not medias:
986 medias = playlist_medias
987
988 entries = []
989 for num, media_meta in enumerate(medias, start=1):
990 formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
991 if not formats:
992 continue
993 self._sort_formats(formats)
994
995 video_id = media_meta.get('externalId')
996 if not video_id:
997 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
998
999 title = media_meta.get('caption')
1000 if not title:
1001 title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
1002
1003 duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
1004
1005 images = []
1006 for image in media_meta.get('images', {}).values():
1007 images.extend(image.values())
1008 if 'image' in media_meta:
1009 images.append(media_meta['image'])
1010
1011 thumbnails = [{
1012 'url': image.get('href'),
1013 'width': int_or_none(image.get('width')),
1014 'height': int_or_none(image.get('height')),
1015 } for image in images]
1016
1017 entries.append({
1018 'id': video_id,
1019 'title': title,
1020 'thumbnails': thumbnails,
1021 'duration': duration,
1022 'timestamp': timestamp,
1023 'formats': formats,
1024 'subtitles': subtitles,
1025 })
1026
1027 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
1028
1029
1030 class BBCCoUkArticleIE(InfoExtractor):
1031 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
1032 IE_NAME = 'bbc.co.uk:article'
1033 IE_DESC = 'BBC articles'
1034
1035 _TEST = {
1036 'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer',
1037 'info_dict': {
1038 'id': '3jNQLTMrPlYGTBn0WV6M2MS',
1039 'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four',
1040 'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.',
1041 },
1042 'playlist_count': 4,
1043 'add_ie': ['BBCCoUk'],
1044 }
1045
1046 def _real_extract(self, url):
1047 playlist_id = self._match_id(url)
1048
1049 webpage = self._download_webpage(url, playlist_id)
1050
1051 title = self._og_search_title(webpage)
1052 description = self._og_search_description(webpage).strip()
1053
1054 entries = [self.url_result(programme_url) for programme_url in re.findall(
1055 r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
1056
1057 return self.playlist_result(entries, playlist_id, title, description)
1058
1059
1060 class BBCCoUkPlaylistBaseIE(InfoExtractor):
1061 def _entries(self, webpage, url, playlist_id):
1062 single_page = 'page' in compat_urlparse.parse_qs(
1063 compat_urlparse.urlparse(url).query)
1064 for page_num in itertools.count(2):
1065 for video_id in re.findall(
1066 self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage):
1067 yield self.url_result(
1068 self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
1069 if single_page:
1070 return
1071 next_page = self._search_regex(
1072 r'<li[^>]+class=(["\'])pagination_+next\1[^>]*><a[^>]+href=(["\'])(?P<url>(?:(?!\2).)+)\2',
1073 webpage, 'next page url', default=None, group='url')
1074 if not next_page:
1075 break
1076 webpage = self._download_webpage(
1077 compat_urlparse.urljoin(url, next_page), playlist_id,
1078 'Downloading page %d' % page_num, page_num)
1079
1080 def _real_extract(self, url):
1081 playlist_id = self._match_id(url)
1082
1083 webpage = self._download_webpage(url, playlist_id)
1084
1085 title, description = self._extract_title_and_description(webpage)
1086
1087 return self.playlist_result(
1088 self._entries(webpage, url, playlist_id),
1089 playlist_id, title, description)
1090
1091
1092 class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
1093 IE_NAME = 'bbc.co.uk:iplayer:playlist'
1094 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
1095 _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
1096 _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
1097 _TESTS = [{
1098 'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v',
1099 'info_dict': {
1100 'id': 'b05rcz9v',
1101 'title': 'The Disappearance',
1102 'description': 'French thriller serial about a missing teenager.',
1103 },
1104 'playlist_mincount': 6,
1105 'skip': 'This programme is not currently available on BBC iPlayer',
1106 }, {
1107 # Available for over a year unlike 30 days for most other programmes
1108 'url': 'http://www.bbc.co.uk/iplayer/group/p02tcc32',
1109 'info_dict': {
1110 'id': 'p02tcc32',
1111 'title': 'Bohemian Icons',
1112 'description': 'md5:683e901041b2fe9ba596f2ab04c4dbe7',
1113 },
1114 'playlist_mincount': 10,
1115 }]
1116
1117 def _extract_title_and_description(self, webpage):
1118 title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
1119 description = self._search_regex(
1120 r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
1121 webpage, 'description', fatal=False, group='value')
1122 return title, description
1123
1124
1125 class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
1126 IE_NAME = 'bbc.co.uk:playlist'
1127 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
1128 _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
1129 _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
1130 _TESTS = [{
1131 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1132 'info_dict': {
1133 'id': 'b05rcz9v',
1134 'title': 'The Disappearance - Clips - BBC Four',
1135 'description': 'French thriller serial about a missing teenager.',
1136 },
1137 'playlist_mincount': 7,
1138 }, {
1139 # multipage playlist, explicit page
1140 'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips?page=1',
1141 'info_dict': {
1142 'id': 'b00mfl7n',
1143 'title': 'Frozen Planet - Clips - BBC One',
1144 'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1145 },
1146 'playlist_mincount': 24,
1147 }, {
1148 # multipage playlist, all pages
1149 'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips',
1150 'info_dict': {
1151 'id': 'b00mfl7n',
1152 'title': 'Frozen Planet - Clips - BBC One',
1153 'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1154 },
1155 'playlist_mincount': 142,
1156 }, {
1157 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/broadcasts/2016/06',
1158 'only_matching': True,
1159 }, {
1160 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1161 'only_matching': True,
1162 }, {
1163 'url': 'http://www.bbc.co.uk/programmes/b055jkys/episodes/player',
1164 'only_matching': True,
1165 }]
1166
1167 def _extract_title_and_description(self, webpage):
1168 title = self._og_search_title(webpage, fatal=False)
1169 description = self._og_search_description(webpage)
1170 return title, description
|
