]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/npo.py
New upstream version 2017.09.24
[youtubedl] / youtube_dl / extractor / npo.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..compat import (
7 compat_HTTPError,
8 compat_str,
9 )
10 from ..utils import (
11 determine_ext,
12 ExtractorError,
13 fix_xml_ampersands,
14 orderedSet,
15 parse_duration,
16 qualities,
17 strip_jsonp,
18 unified_strdate,
19 )
20
21
22 class NPOBaseIE(InfoExtractor):
23 def _get_token(self, video_id):
24 return self._download_json(
25 'http://ida.omroep.nl/app.php/auth', video_id,
26 note='Downloading token')['token']
27
28
29 class NPOIE(NPOBaseIE):
30 IE_NAME = 'npo'
31 IE_DESC = 'npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl'
32 _VALID_URL = r'''(?x)
33 (?:
34 npo:|
35 https?://
36 (?:www\.)?
37 (?:
38 npo\.nl/(?!(?:live|radio)/)(?:[^/]+/){2}|
39 ntr\.nl/(?:[^/]+/){2,}|
40 omroepwnl\.nl/video/fragment/[^/]+__|
41 (?:zapp|npo3)\.nl/(?:[^/]+/){2}
42 )
43 )
44 (?P<id>[^/?#]+)
45 '''
46
47 _TESTS = [{
48 'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
49 'md5': '4b3f9c429157ec4775f2c9cb7b911016',
50 'info_dict': {
51 'id': 'VPWON_1220719',
52 'ext': 'm4v',
53 'title': 'Nieuwsuur',
54 'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
55 'upload_date': '20140622',
56 },
57 }, {
58 'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
59 'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
60 'info_dict': {
61 'id': 'VARA_101191800',
62 'ext': 'm4v',
63 'title': 'De Mega Mike & Mega Thomas show: The best of.',
64 'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
65 'upload_date': '20090227',
66 'duration': 2400,
67 },
68 }, {
69 'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
70 'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
71 'info_dict': {
72 'id': 'VPWON_1169289',
73 'ext': 'm4v',
74 'title': 'Tegenlicht: Zwart geld. De toekomst komt uit Afrika',
75 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
76 'upload_date': '20130225',
77 'duration': 3000,
78 },
79 }, {
80 'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
81 'info_dict': {
82 'id': 'WO_VPRO_043706',
83 'ext': 'm4v',
84 'title': 'De nieuwe mens - Deel 1',
85 'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
86 'duration': 4680,
87 },
88 'params': {
89 'skip_download': True,
90 }
91 }, {
92 # non asf in streams
93 'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
94 'info_dict': {
95 'id': 'WO_NOS_762771',
96 'ext': 'mp4',
97 'title': 'Hoe gaat Europa verder na Parijs?',
98 },
99 'params': {
100 'skip_download': True,
101 }
102 }, {
103 'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content',
104 'info_dict': {
105 'id': 'VPWON_1233944',
106 'ext': 'm4v',
107 'title': 'Aap, poot, pies',
108 'description': 'md5:c9c8005d1869ae65b858e82c01a91fde',
109 'upload_date': '20150508',
110 'duration': 599,
111 },
112 'params': {
113 'skip_download': True,
114 }
115 }, {
116 'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698',
117 'info_dict': {
118 'id': 'POW_00996502',
119 'ext': 'm4v',
120 'title': '''"Dit is wel een 'landslide'..."''',
121 'description': 'md5:f8d66d537dfb641380226e31ca57b8e8',
122 'upload_date': '20150508',
123 'duration': 462,
124 },
125 'params': {
126 'skip_download': True,
127 }
128 }, {
129 # audio
130 'url': 'http://www.npo.nl/jouw-stad-rotterdam/29-01-2017/RBX_FUNX_6683215/RBX_FUNX_7601437',
131 'info_dict': {
132 'id': 'RBX_FUNX_6683215',
133 'ext': 'mp3',
134 'title': 'Jouw Stad Rotterdam',
135 'description': 'md5:db251505244f097717ec59fabc372d9f',
136 },
137 'params': {
138 'skip_download': True,
139 }
140 }, {
141 'url': 'http://www.zapp.nl/de-bzt-show/gemist/KN_1687547',
142 'only_matching': True,
143 }, {
144 'url': 'http://www.zapp.nl/de-bzt-show/filmpjes/POMS_KN_7315118',
145 'only_matching': True,
146 }, {
147 'url': 'http://www.zapp.nl/beste-vrienden-quiz/extra-video-s/WO_NTR_1067990',
148 'only_matching': True,
149 }, {
150 'url': 'https://www.npo3.nl/3onderzoekt/16-09-2015/VPWON_1239870',
151 'only_matching': True,
152 }, {
153 # live stream
154 'url': 'npo:LI_NL1_4188102',
155 'only_matching': True,
156 }, {
157 'url': 'http://www.npo.nl/radio-gaga/13-06-2017/BNN_101383373',
158 'only_matching': True,
159 }]
160
161 def _real_extract(self, url):
162 video_id = self._match_id(url)
163 return self._get_info(video_id)
164
165 def _get_info(self, video_id):
166 metadata = self._download_json(
167 'http://e.omroep.nl/metadata/%s' % video_id,
168 video_id,
169 # We have to remove the javascript callback
170 transform_source=strip_jsonp,
171 )
172
173 # For some videos actual video id (prid) is different (e.g. for
174 # http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698
175 # video id is POMS_WNL_853698 but prid is POW_00996502)
176 video_id = metadata.get('prid') or video_id
177
178 # titel is too generic in some cases so utilize aflevering_titel as well
179 # when available (e.g. http://tegenlicht.vpro.nl/afleveringen/2014-2015/access-to-africa.html)
180 title = metadata['titel']
181 sub_title = metadata.get('aflevering_titel')
182 if sub_title and sub_title != title:
183 title += ': %s' % sub_title
184
185 token = self._get_token(video_id)
186
187 formats = []
188 urls = set()
189
190 quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std'])
191 items = self._download_json(
192 'http://ida.omroep.nl/app.php/%s' % video_id, video_id,
193 'Downloading formats JSON', query={
194 'adaptive': 'yes',
195 'token': token,
196 })['items'][0]
197 for num, item in enumerate(items):
198 item_url = item.get('url')
199 if not item_url or item_url in urls:
200 continue
201 urls.add(item_url)
202 format_id = self._search_regex(
203 r'video/ida/([^/]+)', item_url, 'format id',
204 default=None)
205
206 def add_format_url(format_url):
207 formats.append({
208 'url': format_url,
209 'format_id': format_id,
210 'quality': quality(format_id),
211 })
212
213 # Example: http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706
214 if item.get('contentType') in ('url', 'audio'):
215 add_format_url(item_url)
216 continue
217
218 try:
219 stream_info = self._download_json(
220 item_url + '&type=json', video_id,
221 'Downloading %s stream JSON'
222 % item.get('label') or item.get('format') or format_id or num)
223 except ExtractorError as ee:
224 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
225 error = (self._parse_json(
226 ee.cause.read().decode(), video_id,
227 fatal=False) or {}).get('errorstring')
228 if error:
229 raise ExtractorError(error, expected=True)
230 raise
231 # Stream URL instead of JSON, example: npo:LI_NL1_4188102
232 if isinstance(stream_info, compat_str):
233 if not stream_info.startswith('http'):
234 continue
235 video_url = stream_info
236 # JSON
237 else:
238 video_url = stream_info.get('url')
239 if not video_url or video_url in urls:
240 continue
241 urls.add(item_url)
242 if determine_ext(video_url) == 'm3u8':
243 formats.extend(self._extract_m3u8_formats(
244 video_url, video_id, ext='mp4',
245 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
246 else:
247 add_format_url(video_url)
248
249 is_live = metadata.get('medium') == 'live'
250
251 if not is_live:
252 for num, stream in enumerate(metadata.get('streams', [])):
253 stream_url = stream.get('url')
254 if not stream_url or stream_url in urls:
255 continue
256 urls.add(stream_url)
257 # smooth streaming is not supported
258 stream_type = stream.get('type', '').lower()
259 if stream_type in ['ss', 'ms']:
260 continue
261 if stream_type == 'hds':
262 f4m_formats = self._extract_f4m_formats(
263 stream_url, video_id, fatal=False)
264 # f4m downloader downloads only piece of live stream
265 for f4m_format in f4m_formats:
266 f4m_format['preference'] = -1
267 formats.extend(f4m_formats)
268 elif stream_type == 'hls':
269 formats.extend(self._extract_m3u8_formats(
270 stream_url, video_id, ext='mp4', fatal=False))
271 # Example: http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706
272 elif '.asf' in stream_url:
273 asx = self._download_xml(
274 stream_url, video_id,
275 'Downloading stream %d ASX playlist' % num,
276 transform_source=fix_xml_ampersands, fatal=False)
277 if not asx:
278 continue
279 ref = asx.find('./ENTRY/Ref')
280 if ref is None:
281 continue
282 video_url = ref.get('href')
283 if not video_url or video_url in urls:
284 continue
285 urls.add(video_url)
286 formats.append({
287 'url': video_url,
288 'ext': stream.get('formaat', 'asf'),
289 'quality': stream.get('kwaliteit'),
290 'preference': -10,
291 })
292 else:
293 formats.append({
294 'url': stream_url,
295 'quality': stream.get('kwaliteit'),
296 })
297
298 self._sort_formats(formats)
299
300 subtitles = {}
301 if metadata.get('tt888') == 'ja':
302 subtitles['nl'] = [{
303 'ext': 'vtt',
304 'url': 'http://tt888.omroep.nl/tt888/%s' % video_id,
305 }]
306
307 return {
308 'id': video_id,
309 'title': self._live_title(title) if is_live else title,
310 'description': metadata.get('info'),
311 'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
312 'upload_date': unified_strdate(metadata.get('gidsdatum')),
313 'duration': parse_duration(metadata.get('tijdsduur')),
314 'formats': formats,
315 'subtitles': subtitles,
316 'is_live': is_live,
317 }
318
319
320 class NPOLiveIE(NPOBaseIE):
321 IE_NAME = 'npo.nl:live'
322 _VALID_URL = r'https?://(?:www\.)?npo\.nl/live(?:/(?P<id>[^/?#&]+))?'
323
324 _TESTS = [{
325 'url': 'http://www.npo.nl/live/npo-1',
326 'info_dict': {
327 'id': 'LI_NL1_4188102',
328 'display_id': 'npo-1',
329 'ext': 'mp4',
330 'title': 're:^NPO 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
331 'is_live': True,
332 },
333 'params': {
334 'skip_download': True,
335 }
336 }, {
337 'url': 'http://www.npo.nl/live',
338 'only_matching': True,
339 }]
340
341 def _real_extract(self, url):
342 display_id = self._match_id(url) or 'npo-1'
343
344 webpage = self._download_webpage(url, display_id)
345
346 live_id = self._search_regex(
347 [r'media-id="([^"]+)"', r'data-prid="([^"]+)"'], webpage, 'live id')
348
349 return {
350 '_type': 'url_transparent',
351 'url': 'npo:%s' % live_id,
352 'ie_key': NPOIE.ie_key(),
353 'id': live_id,
354 'display_id': display_id,
355 }
356
357
358 class NPORadioIE(InfoExtractor):
359 IE_NAME = 'npo.nl:radio'
360 _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
361
362 _TEST = {
363 'url': 'http://www.npo.nl/radio/radio-1',
364 'info_dict': {
365 'id': 'radio-1',
366 'ext': 'mp3',
367 'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
368 'is_live': True,
369 },
370 'params': {
371 'skip_download': True,
372 }
373 }
374
375 @staticmethod
376 def _html_get_attribute_regex(attribute):
377 return r'{0}\s*=\s*\'([^\']+)\''.format(attribute)
378
379 def _real_extract(self, url):
380 video_id = self._match_id(url)
381
382 webpage = self._download_webpage(url, video_id)
383
384 title = self._html_search_regex(
385 self._html_get_attribute_regex('data-channel'), webpage, 'title')
386
387 stream = self._parse_json(
388 self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
389 video_id)
390
391 codec = stream.get('codec')
392
393 return {
394 'id': video_id,
395 'url': stream['url'],
396 'title': self._live_title(title),
397 'acodec': codec,
398 'ext': codec,
399 'is_live': True,
400 }
401
402
403 class NPORadioFragmentIE(InfoExtractor):
404 IE_NAME = 'npo.nl:radio:fragment'
405 _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
406
407 _TEST = {
408 'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
409 'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
410 'info_dict': {
411 'id': '174356',
412 'ext': 'mp3',
413 'title': 'Jubileumconcert Willeke Alberti',
414 },
415 }
416
417 def _real_extract(self, url):
418 audio_id = self._match_id(url)
419
420 webpage = self._download_webpage(url, audio_id)
421
422 title = self._html_search_regex(
423 r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
424 webpage, 'title')
425
426 audio_url = self._search_regex(
427 r"data-streams='([^']+)'", webpage, 'audio url')
428
429 return {
430 'id': audio_id,
431 'url': audio_url,
432 'title': title,
433 }
434
435
436 class NPODataMidEmbedIE(InfoExtractor):
437 def _real_extract(self, url):
438 display_id = self._match_id(url)
439 webpage = self._download_webpage(url, display_id)
440 video_id = self._search_regex(
441 r'data-mid=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video_id', group='id')
442 return {
443 '_type': 'url_transparent',
444 'ie_key': 'NPO',
445 'url': 'npo:%s' % video_id,
446 'display_id': display_id
447 }
448
449
450 class SchoolTVIE(NPODataMidEmbedIE):
451 IE_NAME = 'schooltv'
452 _VALID_URL = r'https?://(?:www\.)?schooltv\.nl/video/(?P<id>[^/?#&]+)'
453
454 _TEST = {
455 'url': 'http://www.schooltv.nl/video/ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam/',
456 'info_dict': {
457 'id': 'WO_NTR_429477',
458 'display_id': 'ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam',
459 'title': 'Ademhaling: De hele dag haal je adem. Maar wat gebeurt er dan eigenlijk in je lichaam?',
460 'ext': 'mp4',
461 'description': 'md5:abfa0ff690adb73fd0297fd033aaa631'
462 },
463 'params': {
464 # Skip because of m3u8 download
465 'skip_download': True
466 }
467 }
468
469
470 class HetKlokhuisIE(NPODataMidEmbedIE):
471 IE_NAME = 'hetklokhuis'
472 _VALID_URL = r'https?://(?:www\.)?hetklokhuis.nl/[^/]+/\d+/(?P<id>[^/?#&]+)'
473
474 _TEST = {
475 'url': 'http://hetklokhuis.nl/tv-uitzending/3471/Zwaartekrachtsgolven',
476 'info_dict': {
477 'id': 'VPWON_1260528',
478 'display_id': 'Zwaartekrachtsgolven',
479 'ext': 'm4v',
480 'title': 'Het Klokhuis: Zwaartekrachtsgolven',
481 'description': 'md5:c94f31fb930d76c2efa4a4a71651dd48',
482 'upload_date': '20170223',
483 },
484 'params': {
485 'skip_download': True
486 }
487 }
488
489
490 class NPOPlaylistBaseIE(NPOIE):
491 def _real_extract(self, url):
492 playlist_id = self._match_id(url)
493
494 webpage = self._download_webpage(url, playlist_id)
495
496 entries = [
497 self.url_result('npo:%s' % video_id if not video_id.startswith('http') else video_id)
498 for video_id in orderedSet(re.findall(self._PLAYLIST_ENTRY_RE, webpage))
499 ]
500
501 playlist_title = self._html_search_regex(
502 self._PLAYLIST_TITLE_RE, webpage, 'playlist title',
503 default=None) or self._og_search_title(webpage)
504
505 return self.playlist_result(entries, playlist_id, playlist_title)
506
507
508 class VPROIE(NPOPlaylistBaseIE):
509 IE_NAME = 'vpro'
510 _VALID_URL = r'https?://(?:www\.)?(?:(?:tegenlicht\.)?vpro|2doc)\.nl/(?:[^/]+/)*(?P<id>[^/]+)\.html'
511 _PLAYLIST_TITLE_RE = (r'<h1[^>]+class=["\'].*?\bmedia-platform-title\b.*?["\'][^>]*>([^<]+)',
512 r'<h5[^>]+class=["\'].*?\bmedia-platform-subtitle\b.*?["\'][^>]*>([^<]+)')
513 _PLAYLIST_ENTRY_RE = r'data-media-id="([^"]+)"'
514
515 _TESTS = [
516 {
517 'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
518 'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
519 'info_dict': {
520 'id': 'VPWON_1169289',
521 'ext': 'm4v',
522 'title': 'De toekomst komt uit Afrika',
523 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
524 'upload_date': '20130225',
525 },
526 'skip': 'Video gone',
527 },
528 {
529 'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html',
530 'info_dict': {
531 'id': 'sergio-herman',
532 'title': 'sergio herman: fucking perfect',
533 },
534 'playlist_count': 2,
535 },
536 {
537 # playlist with youtube embed
538 'url': 'http://www.vpro.nl/programmas/2doc/2015/education-education.html',
539 'info_dict': {
540 'id': 'education-education',
541 'title': 'education education',
542 },
543 'playlist_count': 2,
544 },
545 {
546 'url': 'http://www.2doc.nl/documentaires/series/2doc/2015/oktober/de-tegenprestatie.html',
547 'info_dict': {
548 'id': 'de-tegenprestatie',
549 'title': 'De Tegenprestatie',
550 },
551 'playlist_count': 2,
552 }, {
553 'url': 'http://www.2doc.nl/speel~VARA_101375237~mh17-het-verdriet-van-nederland~.html',
554 'info_dict': {
555 'id': 'VARA_101375237',
556 'ext': 'm4v',
557 'title': 'MH17: Het verdriet van Nederland',
558 'description': 'md5:09e1a37c1fdb144621e22479691a9f18',
559 'upload_date': '20150716',
560 },
561 'params': {
562 # Skip because of m3u8 download
563 'skip_download': True
564 },
565 }
566 ]
567
568
569 class WNLIE(NPOPlaylistBaseIE):
570 IE_NAME = 'wnl'
571 _VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+'
572 _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>'
573 _PLAYLIST_ENTRY_RE = r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>Deel \d+'
574
575 _TESTS = [{
576 'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515',
577 'info_dict': {
578 'id': 'vandaag-de-dag-6-mei',
579 'title': 'Vandaag de Dag 6 mei',
580 },
581 'playlist_count': 4,
582 }]
583
584
585 class AndereTijdenIE(NPOPlaylistBaseIE):
586 IE_NAME = 'anderetijden'
587 _VALID_URL = r'https?://(?:www\.)?anderetijden\.nl/programma/(?:[^/]+/)+(?P<id>[^/?#&]+)'
588 _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class=["\'].*?\bpage-title\b.*?["\'][^>]*>(.+?)</h1>'
589 _PLAYLIST_ENTRY_RE = r'<figure[^>]+class=["\']episode-container episode-page["\'][^>]+data-prid=["\'](.+?)["\']'
590
591 _TESTS = [{
592 'url': 'http://anderetijden.nl/programma/1/Andere-Tijden/aflevering/676/Duitse-soldaten-over-de-Slag-bij-Arnhem',
593 'info_dict': {
594 'id': 'Duitse-soldaten-over-de-Slag-bij-Arnhem',
595 'title': 'Duitse soldaten over de Slag bij Arnhem',
596 },
597 'playlist_count': 3,
598 }]