]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bbc.py
eac9a5a4668961571c9162ca2be93aba3211438a
[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 clean_html,
10 dict_get,
11 ExtractorError,
12 float_or_none,
13 get_element_by_class,
14 int_or_none,
15 js_to_json,
16 parse_duration,
17 parse_iso8601,
18 try_get,
19 unescapeHTML,
20 urlencode_postdata,
21 urljoin,
22 )
23 from ..compat import (
24 compat_HTTPError,
25 compat_urlparse,
26 )
27
28
29 class BBCCoUkIE(InfoExtractor):
30 IE_NAME = 'bbc.co.uk'
31 IE_DESC = 'BBC iPlayer'
32 _ID_REGEX = r'(?:[pbm][\da-z]{7}|w[\da-z]{7,14})'
33 _VALID_URL = r'''(?x)
34 https?://
35 (?:www\.)?bbc\.co\.uk/
36 (?:
37 programmes/(?!articles/)|
38 iplayer(?:/[^/]+)?/(?:episode/|playlist/)|
39 music/(?:clips|audiovideo/popular)[/#]|
40 radio/player/|
41 events/[^/]+/play/[^/]+/
42 )
43 (?P<id>%s)(?!/(?:episodes|broadcasts|clips))
44 ''' % _ID_REGEX
45
46 _LOGIN_URL = 'https://account.bbc.com/signin'
47 _NETRC_MACHINE = 'bbc'
48
49 _MEDIASELECTOR_URLS = [
50 # Provides HQ HLS streams with even better quality that pc mediaset but fails
51 # with geolocation in some cases when it's even not geo restricted at all (e.g.
52 # http://www.bbc.co.uk/programmes/b06bp7lf). Also may fail with selectionunavailable.
53 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
54 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
55 ]
56
57 _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
58 _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
59
60 _NAMESPACES = (
61 _MEDIASELECTION_NS,
62 _EMP_PLAYLIST_NS,
63 )
64
65 _TESTS = [
66 {
67 'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
68 'info_dict': {
69 'id': 'b039d07m',
70 'ext': 'flv',
71 'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
72 'description': 'The Canadian poet and songwriter reflects on his musical career.',
73 },
74 'params': {
75 # rtmp download
76 'skip_download': True,
77 }
78 },
79 {
80 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
81 'info_dict': {
82 'id': 'b00yng1d',
83 'ext': 'flv',
84 'title': 'The Man in Black: Series 3: The Printed Name',
85 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
86 'duration': 1800,
87 },
88 'params': {
89 # rtmp download
90 'skip_download': True,
91 },
92 'skip': 'Episode is no longer available on BBC iPlayer Radio',
93 },
94 {
95 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
96 'info_dict': {
97 'id': 'b00yng1d',
98 'ext': 'flv',
99 'title': 'The Voice UK: Series 3: Blind Auditions 5',
100 'description': 'Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.',
101 'duration': 5100,
102 },
103 'params': {
104 # rtmp download
105 'skip_download': True,
106 },
107 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
108 },
109 {
110 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
111 'info_dict': {
112 'id': 'b03k3pb7',
113 'ext': 'flv',
114 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
115 'description': '2. Invasion',
116 'duration': 3600,
117 },
118 'params': {
119 # rtmp download
120 'skip_download': True,
121 },
122 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
123 }, {
124 'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
125 'info_dict': {
126 'id': 'b04v209v',
127 'ext': 'flv',
128 'title': 'Pete Tong, The Essential New Tune Special',
129 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
130 'duration': 10800,
131 },
132 'params': {
133 # rtmp download
134 'skip_download': True,
135 },
136 'skip': 'Episode is no longer available on BBC iPlayer Radio',
137 }, {
138 'url': 'http://www.bbc.co.uk/music/clips/p022h44b',
139 'note': 'Audio',
140 'info_dict': {
141 'id': 'p022h44j',
142 'ext': 'flv',
143 'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
144 'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
145 'duration': 227,
146 },
147 'params': {
148 # rtmp download
149 'skip_download': True,
150 }
151 }, {
152 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
153 'note': 'Video',
154 'info_dict': {
155 'id': 'p025c103',
156 'ext': 'flv',
157 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
158 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
159 'duration': 226,
160 },
161 'params': {
162 # rtmp download
163 'skip_download': True,
164 }
165 }, {
166 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
167 'info_dict': {
168 'id': 'p02n76xf',
169 'ext': 'flv',
170 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
171 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
172 'duration': 3540,
173 },
174 'params': {
175 # rtmp download
176 'skip_download': True,
177 },
178 'skip': 'geolocation',
179 }, {
180 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
181 'info_dict': {
182 'id': 'b05zmgw1',
183 'ext': 'flv',
184 'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.',
185 'title': 'Royal Academy Summer Exhibition',
186 'duration': 3540,
187 },
188 'params': {
189 # rtmp download
190 'skip_download': True,
191 },
192 'skip': 'geolocation',
193 }, {
194 # iptv-all mediaset fails with geolocation however there is no geo restriction
195 # for this programme at all
196 'url': 'http://www.bbc.co.uk/programmes/b06rkn85',
197 'info_dict': {
198 'id': 'b06rkms3',
199 'ext': 'flv',
200 'title': "Best of the Mini-Mixes 2015: Part 3, Annie Mac's Friday Night - BBC Radio 1",
201 'description': "Annie has part three in the Best of the Mini-Mixes 2015, plus the year's Most Played!",
202 },
203 'params': {
204 # rtmp download
205 'skip_download': True,
206 },
207 'skip': 'Now it\'s really geo-restricted',
208 }, {
209 # compact player (https://github.com/rg3/youtube-dl/issues/8147)
210 'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
211 'info_dict': {
212 'id': 'p028bfkj',
213 'ext': 'flv',
214 'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
215 'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
216 },
217 'params': {
218 # rtmp download
219 'skip_download': True,
220 },
221 }, {
222 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
223 'only_matching': True,
224 }, {
225 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
226 'only_matching': True,
227 }, {
228 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
229 'only_matching': True,
230 }, {
231 'url': 'http://www.bbc.co.uk/radio/player/p03cchwf',
232 'only_matching': True,
233 }, {
234 'url': 'https://www.bbc.co.uk/music/audiovideo/popular#p055bc55',
235 'only_matching': True,
236 }, {
237 'url': 'http://www.bbc.co.uk/programmes/w3csv1y9',
238 'only_matching': True,
239 }, {
240 'url': 'https://www.bbc.co.uk/programmes/m00005xn',
241 'only_matching': True,
242 }, {
243 'url': 'https://www.bbc.co.uk/programmes/w172w4dww1jqt5s',
244 'only_matching': True,
245 }]
246
247 _USP_RE = r'/([^/]+?)\.ism(?:\.hlsv2\.ism)?/[^/]+\.m3u8'
248
249 def _login(self):
250 username, password = self._get_login_info()
251 if username is None:
252 return
253
254 login_page = self._download_webpage(
255 self._LOGIN_URL, None, 'Downloading signin page')
256
257 login_form = self._hidden_inputs(login_page)
258
259 login_form.update({
260 'username': username,
261 'password': password,
262 })
263
264 post_url = urljoin(self._LOGIN_URL, self._search_regex(
265 r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
266 'post url', default=self._LOGIN_URL, group='url'))
267
268 response, urlh = self._download_webpage_handle(
269 post_url, None, 'Logging in', data=urlencode_postdata(login_form),
270 headers={'Referer': self._LOGIN_URL})
271
272 if self._LOGIN_URL in urlh.geturl():
273 error = clean_html(get_element_by_class('form-message', response))
274 if error:
275 raise ExtractorError(
276 'Unable to login: %s' % error, expected=True)
277 raise ExtractorError('Unable to log in')
278
279 def _real_initialize(self):
280 self._login()
281
282 class MediaSelectionError(Exception):
283 def __init__(self, id):
284 self.id = id
285
286 def _extract_asx_playlist(self, connection, programme_id):
287 asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
288 return [ref.get('href') for ref in asx.findall('./Entry/ref')]
289
290 def _extract_items(self, playlist):
291 return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
292
293 def _findall_ns(self, element, xpath):
294 elements = []
295 for ns in self._NAMESPACES:
296 elements.extend(element.findall(xpath % ns))
297 return elements
298
299 def _extract_medias(self, media_selection):
300 error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS)
301 if error is None:
302 media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS)
303 if error is not None:
304 raise BBCCoUkIE.MediaSelectionError(error.get('id'))
305 return self._findall_ns(media_selection, './{%s}media')
306
307 def _extract_connections(self, media):
308 return self._findall_ns(media, './{%s}connection')
309
310 def _get_subtitles(self, media, programme_id):
311 subtitles = {}
312 for connection in self._extract_connections(media):
313 captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
314 lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
315 subtitles[lang] = [
316 {
317 'url': connection.get('href'),
318 'ext': 'ttml',
319 },
320 ]
321 return subtitles
322
323 def _raise_extractor_error(self, media_selection_error):
324 raise ExtractorError(
325 '%s returned error: %s' % (self.IE_NAME, media_selection_error.id),
326 expected=True)
327
328 def _download_media_selector(self, programme_id):
329 last_exception = None
330 for mediaselector_url in self._MEDIASELECTOR_URLS:
331 try:
332 return self._download_media_selector_url(
333 mediaselector_url % programme_id, programme_id)
334 except BBCCoUkIE.MediaSelectionError as e:
335 if e.id in ('notukerror', 'geolocation', 'selectionunavailable'):
336 last_exception = e
337 continue
338 self._raise_extractor_error(e)
339 self._raise_extractor_error(last_exception)
340
341 def _download_media_selector_url(self, url, programme_id=None):
342 media_selection = self._download_xml(
343 url, programme_id, 'Downloading media selection XML',
344 expected_status=(403, 404))
345 return self._process_media_selector(media_selection, programme_id)
346
347 def _process_media_selector(self, media_selection, programme_id):
348 formats = []
349 subtitles = None
350 urls = []
351
352 for media in self._extract_medias(media_selection):
353 kind = media.get('kind')
354 if kind in ('video', 'audio'):
355 bitrate = int_or_none(media.get('bitrate'))
356 encoding = media.get('encoding')
357 service = media.get('service')
358 width = int_or_none(media.get('width'))
359 height = int_or_none(media.get('height'))
360 file_size = int_or_none(media.get('media_file_size'))
361 for connection in self._extract_connections(media):
362 href = connection.get('href')
363 if href in urls:
364 continue
365 if href:
366 urls.append(href)
367 conn_kind = connection.get('kind')
368 protocol = connection.get('protocol')
369 supplier = connection.get('supplier')
370 transfer_format = connection.get('transferFormat')
371 format_id = supplier or conn_kind or protocol
372 if service:
373 format_id = '%s_%s' % (service, format_id)
374 # ASX playlist
375 if supplier == 'asx':
376 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
377 formats.append({
378 'url': ref,
379 'format_id': 'ref%s_%s' % (i, format_id),
380 })
381 elif transfer_format == 'dash':
382 formats.extend(self._extract_mpd_formats(
383 href, programme_id, mpd_id=format_id, fatal=False))
384 elif transfer_format == 'hls':
385 formats.extend(self._extract_m3u8_formats(
386 href, programme_id, ext='mp4', entry_protocol='m3u8_native',
387 m3u8_id=format_id, fatal=False))
388 if re.search(self._USP_RE, href):
389 usp_formats = self._extract_m3u8_formats(
390 re.sub(self._USP_RE, r'/\1.ism/\1.m3u8', href),
391 programme_id, ext='mp4', entry_protocol='m3u8_native',
392 m3u8_id=format_id, fatal=False)
393 for f in usp_formats:
394 if f.get('height') and f['height'] > 720:
395 continue
396 formats.append(f)
397 elif transfer_format == 'hds':
398 formats.extend(self._extract_f4m_formats(
399 href, programme_id, f4m_id=format_id, fatal=False))
400 else:
401 if not service and not supplier and bitrate:
402 format_id += '-%d' % bitrate
403 fmt = {
404 'format_id': format_id,
405 'filesize': file_size,
406 }
407 if kind == 'video':
408 fmt.update({
409 'width': width,
410 'height': height,
411 'tbr': bitrate,
412 'vcodec': encoding,
413 })
414 else:
415 fmt.update({
416 'abr': bitrate,
417 'acodec': encoding,
418 'vcodec': 'none',
419 })
420 if protocol in ('http', 'https'):
421 # Direct link
422 fmt.update({
423 'url': href,
424 })
425 elif protocol == 'rtmp':
426 application = connection.get('application', 'ondemand')
427 auth_string = connection.get('authString')
428 identifier = connection.get('identifier')
429 server = connection.get('server')
430 fmt.update({
431 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
432 'play_path': identifier,
433 'app': '%s?%s' % (application, auth_string),
434 'page_url': 'http://www.bbc.co.uk',
435 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
436 'rtmp_live': False,
437 'ext': 'flv',
438 })
439 else:
440 continue
441 formats.append(fmt)
442 elif kind == 'captions':
443 subtitles = self.extract_subtitles(media, programme_id)
444 return formats, subtitles
445
446 def _download_playlist(self, playlist_id):
447 try:
448 playlist = self._download_json(
449 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
450 playlist_id, 'Downloading playlist JSON')
451
452 version = playlist.get('defaultAvailableVersion')
453 if version:
454 smp_config = version['smpConfig']
455 title = smp_config['title']
456 description = smp_config['summary']
457 for item in smp_config['items']:
458 kind = item['kind']
459 if kind not in ('programme', 'radioProgramme'):
460 continue
461 programme_id = item.get('vpid')
462 duration = int_or_none(item.get('duration'))
463 formats, subtitles = self._download_media_selector(programme_id)
464 return programme_id, title, description, duration, formats, subtitles
465 except ExtractorError as ee:
466 if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
467 raise
468
469 # fallback to legacy playlist
470 return self._process_legacy_playlist(playlist_id)
471
472 def _process_legacy_playlist_url(self, url, display_id):
473 playlist = self._download_legacy_playlist_url(url, display_id)
474 return self._extract_from_legacy_playlist(playlist, display_id)
475
476 def _process_legacy_playlist(self, playlist_id):
477 return self._process_legacy_playlist_url(
478 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
479
480 def _download_legacy_playlist_url(self, url, playlist_id=None):
481 return self._download_xml(
482 url, playlist_id, 'Downloading legacy playlist XML')
483
484 def _extract_from_legacy_playlist(self, playlist, playlist_id):
485 no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS)
486 if no_items is not None:
487 reason = no_items.get('reason')
488 if reason == 'preAvailability':
489 msg = 'Episode %s is not yet available' % playlist_id
490 elif reason == 'postAvailability':
491 msg = 'Episode %s is no longer available' % playlist_id
492 elif reason == 'noMedia':
493 msg = 'Episode %s is not currently available' % playlist_id
494 else:
495 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
496 raise ExtractorError(msg, expected=True)
497
498 for item in self._extract_items(playlist):
499 kind = item.get('kind')
500 if kind not in ('programme', 'radioProgramme'):
501 continue
502 title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
503 description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
504 description = description_el.text if description_el is not None else None
505
506 def get_programme_id(item):
507 def get_from_attributes(item):
508 for p in('identifier', 'group'):
509 value = item.get(p)
510 if value and re.match(r'^[pb][\da-z]{7}$', value):
511 return value
512 get_from_attributes(item)
513 mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS)
514 if mediator is not None:
515 return get_from_attributes(mediator)
516
517 programme_id = get_programme_id(item)
518 duration = int_or_none(item.get('duration'))
519
520 if programme_id:
521 formats, subtitles = self._download_media_selector(programme_id)
522 else:
523 formats, subtitles = self._process_media_selector(item, playlist_id)
524 programme_id = playlist_id
525
526 return programme_id, title, description, duration, formats, subtitles
527
528 def _real_extract(self, url):
529 group_id = self._match_id(url)
530
531 webpage = self._download_webpage(url, group_id, 'Downloading video page')
532
533 error = self._search_regex(
534 r'<div\b[^>]+\bclass=["\']smp__message delta["\'][^>]*>([^<]+)<',
535 webpage, 'error', default=None)
536 if error:
537 raise ExtractorError(error, expected=True)
538
539 programme_id = None
540 duration = None
541
542 tviplayer = self._search_regex(
543 r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
544 webpage, 'player', default=None)
545
546 if tviplayer:
547 player = self._parse_json(tviplayer, group_id).get('player', {})
548 duration = int_or_none(player.get('duration'))
549 programme_id = player.get('vpid')
550
551 if not programme_id:
552 programme_id = self._search_regex(
553 r'"vpid"\s*:\s*"(%s)"' % self._ID_REGEX, webpage, 'vpid', fatal=False, default=None)
554
555 if programme_id:
556 formats, subtitles = self._download_media_selector(programme_id)
557 title = self._og_search_title(webpage, default=None) or self._html_search_regex(
558 (r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>',
559 r'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage, 'title')
560 description = self._search_regex(
561 (r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
562 r'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'),
563 webpage, 'description', default=None)
564 if not description:
565 description = self._html_search_meta('description', webpage)
566 else:
567 programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
568
569 self._sort_formats(formats)
570
571 return {
572 'id': programme_id,
573 'title': title,
574 'description': description,
575 'thumbnail': self._og_search_thumbnail(webpage, default=None),
576 'duration': duration,
577 'formats': formats,
578 'subtitles': subtitles,
579 }
580
581
582 class BBCIE(BBCCoUkIE):
583 IE_NAME = 'bbc'
584 IE_DESC = 'BBC'
585 _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
586
587 _MEDIASELECTOR_URLS = [
588 # Provides HQ HLS streams but fails with geolocation in some cases when it's
589 # even not geo restricted at all
590 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
591 # Provides more formats, namely direct mp4 links, but fails on some videos with
592 # notukerror for non UK (?) users (e.g.
593 # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
594 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
595 # Provides fewer formats, but works everywhere for everybody (hopefully)
596 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
597 ]
598
599 _TESTS = [{
600 # article with multiple videos embedded with data-playable containing vpids
601 'url': 'http://www.bbc.com/news/world-europe-32668511',
602 'info_dict': {
603 'id': 'world-europe-32668511',
604 'title': 'Russia stages massive WW2 parade despite Western boycott',
605 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
606 },
607 'playlist_count': 2,
608 }, {
609 # article with multiple videos embedded with data-playable (more videos)
610 'url': 'http://www.bbc.com/news/business-28299555',
611 'info_dict': {
612 'id': 'business-28299555',
613 'title': 'Farnborough Airshow: Video highlights',
614 'description': 'BBC reports and video highlights at the Farnborough Airshow.',
615 },
616 'playlist_count': 9,
617 'skip': 'Save time',
618 }, {
619 # article with multiple videos embedded with `new SMP()`
620 # broken
621 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
622 'info_dict': {
623 'id': '3662a707-0af9-3149-963f-47bea720b460',
624 'title': 'BUGGER',
625 },
626 'playlist_count': 18,
627 }, {
628 # single video embedded with data-playable containing vpid
629 'url': 'http://www.bbc.com/news/world-europe-32041533',
630 'info_dict': {
631 'id': 'p02mprgb',
632 'ext': 'mp4',
633 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
634 'description': 'md5:2868290467291b37feda7863f7a83f54',
635 'duration': 47,
636 'timestamp': 1427219242,
637 'upload_date': '20150324',
638 },
639 'params': {
640 # rtmp download
641 'skip_download': True,
642 }
643 }, {
644 # article with single video embedded with data-playable containing XML playlist
645 # with direct video links as progressiveDownloadUrl (for now these are extracted)
646 # and playlist with f4m and m3u8 as streamingUrl
647 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
648 'info_dict': {
649 'id': '150615_telabyad_kentin_cogu',
650 'ext': 'mp4',
651 'title': "YPG: Tel Abyad'ın tamamı kontrolĆ¼mĆ¼zde",
652 'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
653 'timestamp': 1434397334,
654 'upload_date': '20150615',
655 },
656 'params': {
657 'skip_download': True,
658 }
659 }, {
660 # single video embedded with data-playable containing XML playlists (regional section)
661 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
662 'info_dict': {
663 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
664 'ext': 'mp4',
665 'title': 'Honduras militariza sus hospitales por nuevo escĆ”ndalo de corrupciĆ³n',
666 'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
667 'timestamp': 1434713142,
668 'upload_date': '20150619',
669 },
670 'params': {
671 'skip_download': True,
672 }
673 }, {
674 # single video from video playlist embedded with vxp-playlist-data JSON
675 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
676 'info_dict': {
677 'id': 'p02w6qjc',
678 'ext': 'mp4',
679 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
680 'duration': 56,
681 'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
682 },
683 'params': {
684 'skip_download': True,
685 }
686 }, {
687 # single video story with digitalData
688 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
689 'info_dict': {
690 'id': 'p02q6gc4',
691 'ext': 'flv',
692 'title': 'Sri Lankaā€™s spicy secret',
693 'description': 'As a new train line to Jaffna opens up the countryā€™s north, travellers can experience a truly distinct slice of Tamil culture.',
694 'timestamp': 1437674293,
695 'upload_date': '20150723',
696 },
697 'params': {
698 # rtmp download
699 'skip_download': True,
700 }
701 }, {
702 # single video story without digitalData
703 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
704 'info_dict': {
705 'id': 'p018zqqg',
706 'ext': 'mp4',
707 'title': 'Hyundai Santa Fe Sport: Rock star',
708 'description': 'md5:b042a26142c4154a6e472933cf20793d',
709 'timestamp': 1415867444,
710 'upload_date': '20141113',
711 },
712 'params': {
713 # rtmp download
714 'skip_download': True,
715 }
716 }, {
717 # single video embedded with Morph
718 'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
719 'info_dict': {
720 'id': 'p041vhd0',
721 'ext': 'mp4',
722 'title': "Nigeria v Japan - Men's First Round",
723 'description': 'Live coverage of the first round from Group B at the Amazonia Arena.',
724 'duration': 7980,
725 'uploader': 'BBC Sport',
726 'uploader_id': 'bbc_sport',
727 },
728 'params': {
729 # m3u8 download
730 'skip_download': True,
731 },
732 'skip': 'Georestricted to UK',
733 }, {
734 # single video with playlist.sxml URL in playlist param
735 'url': 'http://www.bbc.com/sport/0/football/33653409',
736 'info_dict': {
737 'id': 'p02xycnp',
738 'ext': 'mp4',
739 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
740 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
741 'duration': 140,
742 },
743 'params': {
744 # rtmp download
745 'skip_download': True,
746 }
747 }, {
748 # article with multiple videos embedded with playlist.sxml in playlist param
749 'url': 'http://www.bbc.com/sport/0/football/34475836',
750 'info_dict': {
751 'id': '34475836',
752 'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
753 'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
754 },
755 'playlist_count': 3,
756 }, {
757 # school report article with single video
758 'url': 'http://www.bbc.co.uk/schoolreport/35744779',
759 'info_dict': {
760 'id': '35744779',
761 'title': 'School which breaks down barriers in Jerusalem',
762 },
763 'playlist_count': 1,
764 }, {
765 # single video with playlist URL from weather section
766 'url': 'http://www.bbc.com/weather/features/33601775',
767 'only_matching': True,
768 }, {
769 # custom redirection to www.bbc.com
770 'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
771 'only_matching': True,
772 }, {
773 # single video article embedded with data-media-vpid
774 'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
775 'only_matching': True,
776 }, {
777 'url': 'https://www.bbc.co.uk/bbcthree/clip/73d0bbd0-abc3-4cea-b3c0-cdae21905eb1',
778 'info_dict': {
779 'id': 'p06556y7',
780 'ext': 'mp4',
781 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
782 'description': 'md5:4b7dfd063d5a789a1512e99662be3ddd',
783 },
784 'params': {
785 'skip_download': True,
786 }
787 }, {
788 # window.__PRELOADED_STATE__
789 'url': 'https://www.bbc.co.uk/radio/play/b0b9z4yl',
790 'info_dict': {
791 'id': 'b0b9z4vz',
792 'ext': 'mp4',
793 'title': 'Prom 6: An American in Paris and Turangalila',
794 'description': 'md5:51cf7d6f5c8553f197e58203bc78dff8',
795 'uploader': 'Radio 3',
796 'uploader_id': 'bbc_radio_three',
797 },
798 }, {
799 'url': 'http://www.bbc.co.uk/learningenglish/chinese/features/lingohack/ep-181227',
800 'info_dict': {
801 'id': 'p06w9tws',
802 'ext': 'mp4',
803 'title': 'md5:2fabf12a726603193a2879a055f72514',
804 'description': 'Learn English words and phrases from this story',
805 },
806 'add_ie': [BBCCoUkIE.ie_key()],
807 }]
808
809 @classmethod
810 def suitable(cls, url):
811 EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE)
812 return (False if any(ie.suitable(url) for ie in EXCLUDE_IE)
813 else super(BBCIE, cls).suitable(url))
814
815 def _extract_from_media_meta(self, media_meta, video_id):
816 # Direct links to media in media metadata (e.g.
817 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
818 # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
819 source_files = media_meta.get('sourceFiles')
820 if source_files:
821 return [{
822 'url': f['url'],
823 'format_id': format_id,
824 'ext': f.get('encoding'),
825 'tbr': float_or_none(f.get('bitrate'), 1000),
826 'filesize': int_or_none(f.get('filesize')),
827 } for format_id, f in source_files.items() if f.get('url')], []
828
829 programme_id = media_meta.get('externalId')
830 if programme_id:
831 return self._download_media_selector(programme_id)
832
833 # Process playlist.sxml as legacy playlist
834 href = media_meta.get('href')
835 if href:
836 playlist = self._download_legacy_playlist_url(href)
837 _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
838 return formats, subtitles
839
840 return [], []
841
842 def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
843 programme_id, title, description, duration, formats, subtitles = \
844 self._process_legacy_playlist_url(url, playlist_id)
845 self._sort_formats(formats)
846 return {
847 'id': programme_id,
848 'title': title,
849 'description': description,
850 'duration': duration,
851 'timestamp': timestamp,
852 'formats': formats,
853 'subtitles': subtitles,
854 }
855
856 def _real_extract(self, url):
857 playlist_id = self._match_id(url)
858
859 webpage = self._download_webpage(url, playlist_id)
860
861 json_ld_info = self._search_json_ld(webpage, playlist_id, default={})
862 timestamp = json_ld_info.get('timestamp')
863
864 playlist_title = json_ld_info.get('title')
865 if not playlist_title:
866 playlist_title = self._og_search_title(
867 webpage, default=None) or self._html_search_regex(
868 r'<title>(.+?)</title>', webpage, 'playlist title', default=None)
869 if playlist_title:
870 playlist_title = re.sub(r'(.+)\s*-\s*BBC.*?$', r'\1', playlist_title).strip()
871
872 playlist_description = json_ld_info.get(
873 'description') or self._og_search_description(webpage, default=None)
874
875 if not timestamp:
876 timestamp = parse_iso8601(self._search_regex(
877 [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
878 r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
879 r'"datePublished":\s*"([^"]+)'],
880 webpage, 'date', default=None))
881
882 entries = []
883
884 # article with multiple videos embedded with playlist.sxml (e.g.
885 # http://www.bbc.com/sport/0/football/34475836)
886 playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
887 playlists.extend(re.findall(r'data-media-id="([^"]+/playlist\.sxml)"', webpage))
888 if playlists:
889 entries = [
890 self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
891 for playlist_url in playlists]
892
893 # news article with multiple videos embedded with data-playable
894 data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
895 if data_playables:
896 for _, data_playable_json in data_playables:
897 data_playable = self._parse_json(
898 unescapeHTML(data_playable_json), playlist_id, fatal=False)
899 if not data_playable:
900 continue
901 settings = data_playable.get('settings', {})
902 if settings:
903 # data-playable with video vpid in settings.playlistObject.items (e.g.
904 # http://www.bbc.com/news/world-us-canada-34473351)
905 playlist_object = settings.get('playlistObject', {})
906 if playlist_object:
907 items = playlist_object.get('items')
908 if items and isinstance(items, list):
909 title = playlist_object['title']
910 description = playlist_object.get('summary')
911 duration = int_or_none(items[0].get('duration'))
912 programme_id = items[0].get('vpid')
913 formats, subtitles = self._download_media_selector(programme_id)
914 self._sort_formats(formats)
915 entries.append({
916 'id': programme_id,
917 'title': title,
918 'description': description,
919 'timestamp': timestamp,
920 'duration': duration,
921 'formats': formats,
922 'subtitles': subtitles,
923 })
924 else:
925 # data-playable without vpid but with a playlist.sxml URLs
926 # in otherSettings.playlist (e.g.
927 # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
928 playlist = data_playable.get('otherSettings', {}).get('playlist', {})
929 if playlist:
930 entry = None
931 for key in ('streaming', 'progressiveDownload'):
932 playlist_url = playlist.get('%sUrl' % key)
933 if not playlist_url:
934 continue
935 try:
936 info = self._extract_from_playlist_sxml(
937 playlist_url, playlist_id, timestamp)
938 if not entry:
939 entry = info
940 else:
941 entry['title'] = info['title']
942 entry['formats'].extend(info['formats'])
943 except Exception as e:
944 # Some playlist URL may fail with 500, at the same time
945 # the other one may work fine (e.g.
946 # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
947 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
948 continue
949 raise
950 if entry:
951 self._sort_formats(entry['formats'])
952 entries.append(entry)
953
954 if entries:
955 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
956
957 # http://www.bbc.co.uk/learningenglish/chinese/features/lingohack/ep-181227
958 group_id = self._search_regex(
959 r'<div[^>]+\bclass=["\']video["\'][^>]+\bdata-pid=["\'](%s)' % self._ID_REGEX,
960 webpage, 'group id', default=None)
961 if playlist_id:
962 return self.url_result(
963 'https://www.bbc.co.uk/programmes/%s' % group_id,
964 ie=BBCCoUkIE.ie_key())
965
966 # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
967 programme_id = self._search_regex(
968 [r'data-(?:video-player|media)-vpid="(%s)"' % self._ID_REGEX,
969 r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
970 r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
971 webpage, 'vpid', default=None)
972
973 if programme_id:
974 formats, subtitles = self._download_media_selector(programme_id)
975 self._sort_formats(formats)
976 # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
977 digital_data = self._parse_json(
978 self._search_regex(
979 r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
980 programme_id, fatal=False)
981 page_info = digital_data.get('page', {}).get('pageInfo', {})
982 title = page_info.get('pageName') or self._og_search_title(webpage)
983 description = page_info.get('description') or self._og_search_description(webpage)
984 timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
985 return {
986 'id': programme_id,
987 'title': title,
988 'description': description,
989 'timestamp': timestamp,
990 'formats': formats,
991 'subtitles': subtitles,
992 }
993
994 # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
995 # There are several setPayload calls may be present but the video
996 # seems to be always related to the first one
997 morph_payload = self._parse_json(
998 self._search_regex(
999 r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
1000 webpage, 'morph payload', default='{}'),
1001 playlist_id, fatal=False)
1002 if morph_payload:
1003 components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
1004 for component in components:
1005 if not isinstance(component, dict):
1006 continue
1007 lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
1008 if not lead_media:
1009 continue
1010 identifiers = lead_media.get('identifiers')
1011 if not identifiers or not isinstance(identifiers, dict):
1012 continue
1013 programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
1014 if not programme_id:
1015 continue
1016 title = lead_media.get('title') or self._og_search_title(webpage)
1017 formats, subtitles = self._download_media_selector(programme_id)
1018 self._sort_formats(formats)
1019 description = lead_media.get('summary')
1020 uploader = lead_media.get('masterBrand')
1021 uploader_id = lead_media.get('mid')
1022 duration = None
1023 duration_d = lead_media.get('duration')
1024 if isinstance(duration_d, dict):
1025 duration = parse_duration(dict_get(
1026 duration_d, ('rawDuration', 'formattedDuration', 'spokenDuration')))
1027 return {
1028 'id': programme_id,
1029 'title': title,
1030 'description': description,
1031 'duration': duration,
1032 'uploader': uploader,
1033 'uploader_id': uploader_id,
1034 'formats': formats,
1035 'subtitles': subtitles,
1036 }
1037
1038 preload_state = self._parse_json(self._search_regex(
1039 r'window\.__PRELOADED_STATE__\s*=\s*({.+?});', webpage,
1040 'preload state', default='{}'), playlist_id, fatal=False)
1041 if preload_state:
1042 current_programme = preload_state.get('programmes', {}).get('current') or {}
1043 programme_id = current_programme.get('id')
1044 if current_programme and programme_id and current_programme.get('type') == 'playable_item':
1045 title = current_programme.get('titles', {}).get('tertiary') or playlist_title
1046 formats, subtitles = self._download_media_selector(programme_id)
1047 self._sort_formats(formats)
1048 synopses = current_programme.get('synopses') or {}
1049 network = current_programme.get('network') or {}
1050 duration = int_or_none(
1051 current_programme.get('duration', {}).get('value'))
1052 thumbnail = None
1053 image_url = current_programme.get('image_url')
1054 if image_url:
1055 thumbnail = image_url.replace('{recipe}', '1920x1920')
1056 return {
1057 'id': programme_id,
1058 'title': title,
1059 'description': dict_get(synopses, ('long', 'medium', 'short')),
1060 'thumbnail': thumbnail,
1061 'duration': duration,
1062 'uploader': network.get('short_title'),
1063 'uploader_id': network.get('id'),
1064 'formats': formats,
1065 'subtitles': subtitles,
1066 }
1067
1068 bbc3_config = self._parse_json(
1069 self._search_regex(
1070 r'(?s)bbcthreeConfig\s*=\s*({.+?})\s*;\s*<', webpage,
1071 'bbcthree config', default='{}'),
1072 playlist_id, transform_source=js_to_json, fatal=False)
1073 if bbc3_config:
1074 bbc3_playlist = try_get(
1075 bbc3_config, lambda x: x['payload']['content']['bbcMedia']['playlist'],
1076 dict)
1077 if bbc3_playlist:
1078 playlist_title = bbc3_playlist.get('title') or playlist_title
1079 thumbnail = bbc3_playlist.get('holdingImageURL')
1080 entries = []
1081 for bbc3_item in bbc3_playlist['items']:
1082 programme_id = bbc3_item.get('versionID')
1083 if not programme_id:
1084 continue
1085 formats, subtitles = self._download_media_selector(programme_id)
1086 self._sort_formats(formats)
1087 entries.append({
1088 'id': programme_id,
1089 'title': playlist_title,
1090 'thumbnail': thumbnail,
1091 'timestamp': timestamp,
1092 'formats': formats,
1093 'subtitles': subtitles,
1094 })
1095 return self.playlist_result(
1096 entries, playlist_id, playlist_title, playlist_description)
1097
1098 def extract_all(pattern):
1099 return list(filter(None, map(
1100 lambda s: self._parse_json(s, playlist_id, fatal=False),
1101 re.findall(pattern, webpage))))
1102
1103 # Multiple video article (e.g.
1104 # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
1105 EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX
1106 entries = []
1107 for match in extract_all(r'new\s+SMP\(({.+?})\)'):
1108 embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
1109 if embed_url and re.match(EMBED_URL, embed_url):
1110 entries.append(embed_url)
1111 entries.extend(re.findall(
1112 r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
1113 if entries:
1114 return self.playlist_result(
1115 [self.url_result(entry_, 'BBCCoUk') for entry_ in entries],
1116 playlist_id, playlist_title, playlist_description)
1117
1118 # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
1119 medias = extract_all(r"data-media-meta='({[^']+})'")
1120
1121 if not medias:
1122 # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
1123 media_asset = self._search_regex(
1124 r'mediaAssetPage\.init\(\s*({.+?}), "/',
1125 webpage, 'media asset', default=None)
1126 if media_asset:
1127 media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
1128 medias = []
1129 for video in media_asset_page.get('videos', {}).values():
1130 medias.extend(video.values())
1131
1132 if not medias:
1133 # Multiple video playlist with single `now playing` entry (e.g.
1134 # http://www.bbc.com/news/video_and_audio/must_see/33767813)
1135 vxp_playlist = self._parse_json(
1136 self._search_regex(
1137 r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>',
1138 webpage, 'playlist data'),
1139 playlist_id)
1140 playlist_medias = []
1141 for item in vxp_playlist:
1142 media = item.get('media')
1143 if not media:
1144 continue
1145 playlist_medias.append(media)
1146 # Download single video if found media with asset id matching the video id from URL
1147 if item.get('advert', {}).get('assetId') == playlist_id:
1148 medias = [media]
1149 break
1150 # Fallback to the whole playlist
1151 if not medias:
1152 medias = playlist_medias
1153
1154 entries = []
1155 for num, media_meta in enumerate(medias, start=1):
1156 formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
1157 if not formats:
1158 continue
1159 self._sort_formats(formats)
1160
1161 video_id = media_meta.get('externalId')
1162 if not video_id:
1163 video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
1164
1165 title = media_meta.get('caption')
1166 if not title:
1167 title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
1168
1169 duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
1170
1171 images = []
1172 for image in media_meta.get('images', {}).values():
1173 images.extend(image.values())
1174 if 'image' in media_meta:
1175 images.append(media_meta['image'])
1176
1177 thumbnails = [{
1178 'url': image.get('href'),
1179 'width': int_or_none(image.get('width')),
1180 'height': int_or_none(image.get('height')),
1181 } for image in images]
1182
1183 entries.append({
1184 'id': video_id,
1185 'title': title,
1186 'thumbnails': thumbnails,
1187 'duration': duration,
1188 'timestamp': timestamp,
1189 'formats': formats,
1190 'subtitles': subtitles,
1191 })
1192
1193 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
1194
1195
1196 class BBCCoUkArticleIE(InfoExtractor):
1197 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
1198 IE_NAME = 'bbc.co.uk:article'
1199 IE_DESC = 'BBC articles'
1200
1201 _TEST = {
1202 'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer',
1203 'info_dict': {
1204 'id': '3jNQLTMrPlYGTBn0WV6M2MS',
1205 'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four',
1206 'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.',
1207 },
1208 'playlist_count': 4,
1209 'add_ie': ['BBCCoUk'],
1210 }
1211
1212 def _real_extract(self, url):
1213 playlist_id = self._match_id(url)
1214
1215 webpage = self._download_webpage(url, playlist_id)
1216
1217 title = self._og_search_title(webpage)
1218 description = self._og_search_description(webpage).strip()
1219
1220 entries = [self.url_result(programme_url) for programme_url in re.findall(
1221 r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
1222
1223 return self.playlist_result(entries, playlist_id, title, description)
1224
1225
1226 class BBCCoUkPlaylistBaseIE(InfoExtractor):
1227 def _entries(self, webpage, url, playlist_id):
1228 single_page = 'page' in compat_urlparse.parse_qs(
1229 compat_urlparse.urlparse(url).query)
1230 for page_num in itertools.count(2):
1231 for video_id in re.findall(
1232 self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage):
1233 yield self.url_result(
1234 self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
1235 if single_page:
1236 return
1237 next_page = self._search_regex(
1238 r'<li[^>]+class=(["\'])pagination_+next\1[^>]*><a[^>]+href=(["\'])(?P<url>(?:(?!\2).)+)\2',
1239 webpage, 'next page url', default=None, group='url')
1240 if not next_page:
1241 break
1242 webpage = self._download_webpage(
1243 compat_urlparse.urljoin(url, next_page), playlist_id,
1244 'Downloading page %d' % page_num, page_num)
1245
1246 def _real_extract(self, url):
1247 playlist_id = self._match_id(url)
1248
1249 webpage = self._download_webpage(url, playlist_id)
1250
1251 title, description = self._extract_title_and_description(webpage)
1252
1253 return self.playlist_result(
1254 self._entries(webpage, url, playlist_id),
1255 playlist_id, title, description)
1256
1257
1258 class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
1259 IE_NAME = 'bbc.co.uk:iplayer:playlist'
1260 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
1261 _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
1262 _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
1263 _TESTS = [{
1264 'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v',
1265 'info_dict': {
1266 'id': 'b05rcz9v',
1267 'title': 'The Disappearance',
1268 'description': 'French thriller serial about a missing teenager.',
1269 },
1270 'playlist_mincount': 6,
1271 'skip': 'This programme is not currently available on BBC iPlayer',
1272 }, {
1273 # Available for over a year unlike 30 days for most other programmes
1274 'url': 'http://www.bbc.co.uk/iplayer/group/p02tcc32',
1275 'info_dict': {
1276 'id': 'p02tcc32',
1277 'title': 'Bohemian Icons',
1278 'description': 'md5:683e901041b2fe9ba596f2ab04c4dbe7',
1279 },
1280 'playlist_mincount': 10,
1281 }]
1282
1283 def _extract_title_and_description(self, webpage):
1284 title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
1285 description = self._search_regex(
1286 r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
1287 webpage, 'description', fatal=False, group='value')
1288 return title, description
1289
1290
1291 class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
1292 IE_NAME = 'bbc.co.uk:playlist'
1293 _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
1294 _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
1295 _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
1296 _TESTS = [{
1297 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1298 'info_dict': {
1299 'id': 'b05rcz9v',
1300 'title': 'The Disappearance - Clips - BBC Four',
1301 'description': 'French thriller serial about a missing teenager.',
1302 },
1303 'playlist_mincount': 7,
1304 }, {
1305 # multipage playlist, explicit page
1306 'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips?page=1',
1307 'info_dict': {
1308 'id': 'b00mfl7n',
1309 'title': 'Frozen Planet - Clips - BBC One',
1310 'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1311 },
1312 'playlist_mincount': 24,
1313 }, {
1314 # multipage playlist, all pages
1315 'url': 'http://www.bbc.co.uk/programmes/b00mfl7n/clips',
1316 'info_dict': {
1317 'id': 'b00mfl7n',
1318 'title': 'Frozen Planet - Clips - BBC One',
1319 'description': 'md5:65dcbf591ae628dafe32aa6c4a4a0d8c',
1320 },
1321 'playlist_mincount': 142,
1322 }, {
1323 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/broadcasts/2016/06',
1324 'only_matching': True,
1325 }, {
1326 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
1327 'only_matching': True,
1328 }, {
1329 'url': 'http://www.bbc.co.uk/programmes/b055jkys/episodes/player',
1330 'only_matching': True,
1331 }]
1332
1333 def _extract_title_and_description(self, webpage):
1334 title = self._og_search_title(webpage, fatal=False)
1335 description = self._og_search_description(webpage)
1336 return title, description