]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/npo.py
New upstream version 2018.03.14
[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 int_or_none,
15 orderedSet,
16 parse_duration,
17 qualities,
18 strip_jsonp,
19 unified_strdate,
20 )
21
22
23 class NPOBaseIE(InfoExtractor):
24 def _get_token(self, video_id):
25 return self._download_json(
26 'http://ida.omroep.nl/app.php/auth', video_id,
27 note='Downloading token')['token']
28
29
30 class NPOIE(NPOBaseIE):
31 IE_NAME = 'npo'
32 IE_DESC = 'npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl'
33 _VALID_URL = r'''(?x)
34 (?:
35 npo:|
36 https?://
37 (?:www\.)?
38 (?:
39 npo\.nl/(?!(?:live|radio)/)(?:[^/]+/){2}|
40 ntr\.nl/(?:[^/]+/){2,}|
41 omroepwnl\.nl/video/fragment/[^/]+__|
42 (?:zapp|npo3)\.nl/(?:[^/]+/){2,}
43 )
44 )
45 (?P<id>[^/?#]+)
46 '''
47
48 _TESTS = [{
49 'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
50 'md5': '4b3f9c429157ec4775f2c9cb7b911016',
51 'info_dict': {
52 'id': 'VPWON_1220719',
53 'ext': 'm4v',
54 'title': 'Nieuwsuur',
55 'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
56 'upload_date': '20140622',
57 },
58 }, {
59 'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
60 'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
61 'info_dict': {
62 'id': 'VARA_101191800',
63 'ext': 'm4v',
64 'title': 'De Mega Mike & Mega Thomas show: The best of.',
65 'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
66 'upload_date': '20090227',
67 'duration': 2400,
68 },
69 }, {
70 'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
71 'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
72 'info_dict': {
73 'id': 'VPWON_1169289',
74 'ext': 'm4v',
75 'title': 'Tegenlicht: Zwart geld. De toekomst komt uit Afrika',
76 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
77 'upload_date': '20130225',
78 'duration': 3000,
79 },
80 }, {
81 'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
82 'info_dict': {
83 'id': 'WO_VPRO_043706',
84 'ext': 'm4v',
85 'title': 'De nieuwe mens - Deel 1',
86 'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
87 'duration': 4680,
88 },
89 'params': {
90 'skip_download': True,
91 }
92 }, {
93 # non asf in streams
94 'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
95 'info_dict': {
96 'id': 'WO_NOS_762771',
97 'ext': 'mp4',
98 'title': 'Hoe gaat Europa verder na Parijs?',
99 },
100 'params': {
101 'skip_download': True,
102 }
103 }, {
104 'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content',
105 'info_dict': {
106 'id': 'VPWON_1233944',
107 'ext': 'm4v',
108 'title': 'Aap, poot, pies',
109 'description': 'md5:c9c8005d1869ae65b858e82c01a91fde',
110 'upload_date': '20150508',
111 'duration': 599,
112 },
113 'params': {
114 'skip_download': True,
115 }
116 }, {
117 'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698',
118 'info_dict': {
119 'id': 'POW_00996502',
120 'ext': 'm4v',
121 'title': '''"Dit is wel een 'landslide'..."''',
122 'description': 'md5:f8d66d537dfb641380226e31ca57b8e8',
123 'upload_date': '20150508',
124 'duration': 462,
125 },
126 'params': {
127 'skip_download': True,
128 }
129 }, {
130 # audio
131 'url': 'http://www.npo.nl/jouw-stad-rotterdam/29-01-2017/RBX_FUNX_6683215/RBX_FUNX_7601437',
132 'info_dict': {
133 'id': 'RBX_FUNX_6683215',
134 'ext': 'mp3',
135 'title': 'Jouw Stad Rotterdam',
136 'description': 'md5:db251505244f097717ec59fabc372d9f',
137 },
138 'params': {
139 'skip_download': True,
140 }
141 }, {
142 'url': 'http://www.zapp.nl/de-bzt-show/gemist/KN_1687547',
143 'only_matching': True,
144 }, {
145 'url': 'http://www.zapp.nl/de-bzt-show/filmpjes/POMS_KN_7315118',
146 'only_matching': True,
147 }, {
148 'url': 'http://www.zapp.nl/beste-vrienden-quiz/extra-video-s/WO_NTR_1067990',
149 'only_matching': True,
150 }, {
151 'url': 'https://www.npo3.nl/3onderzoekt/16-09-2015/VPWON_1239870',
152 'only_matching': True,
153 }, {
154 # live stream
155 'url': 'npo:LI_NL1_4188102',
156 'only_matching': True,
157 }, {
158 'url': 'http://www.npo.nl/radio-gaga/13-06-2017/BNN_101383373',
159 'only_matching': True,
160 }, {
161 'url': 'https://www.zapp.nl/1803-skelterlab/instructie-video-s/740-instructievideo-s/POMS_AT_11736927',
162 'only_matching': True,
163 }]
164
165 def _real_extract(self, url):
166 video_id = self._match_id(url)
167 return self._get_info(video_id)
168
169 def _get_info(self, video_id):
170 metadata = self._download_json(
171 'http://e.omroep.nl/metadata/%s' % video_id,
172 video_id,
173 # We have to remove the javascript callback
174 transform_source=strip_jsonp,
175 )
176
177 error = metadata.get('error')
178 if error:
179 raise ExtractorError(error, expected=True)
180
181 # For some videos actual video id (prid) is different (e.g. for
182 # http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698
183 # video id is POMS_WNL_853698 but prid is POW_00996502)
184 video_id = metadata.get('prid') or video_id
185
186 # titel is too generic in some cases so utilize aflevering_titel as well
187 # when available (e.g. http://tegenlicht.vpro.nl/afleveringen/2014-2015/access-to-africa.html)
188 title = metadata['titel']
189 sub_title = metadata.get('aflevering_titel')
190 if sub_title and sub_title != title:
191 title += ': %s' % sub_title
192
193 token = self._get_token(video_id)
194
195 formats = []
196 urls = set()
197
198 def is_legal_url(format_url):
199 return format_url and format_url not in urls and re.match(
200 r'^(?:https?:)?//', format_url)
201
202 QUALITY_LABELS = ('Laag', 'Normaal', 'Hoog')
203 QUALITY_FORMATS = ('adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std')
204
205 quality_from_label = qualities(QUALITY_LABELS)
206 quality_from_format_id = qualities(QUALITY_FORMATS)
207 items = self._download_json(
208 'http://ida.omroep.nl/app.php/%s' % video_id, video_id,
209 'Downloading formats JSON', query={
210 'adaptive': 'yes',
211 'token': token,
212 })['items'][0]
213 for num, item in enumerate(items):
214 item_url = item.get('url')
215 if not is_legal_url(item_url):
216 continue
217 urls.add(item_url)
218 format_id = self._search_regex(
219 r'video/ida/([^/]+)', item_url, 'format id',
220 default=None)
221
222 item_label = item.get('label')
223
224 def add_format_url(format_url):
225 width = int_or_none(self._search_regex(
226 r'(\d+)[xX]\d+', format_url, 'width', default=None))
227 height = int_or_none(self._search_regex(
228 r'\d+[xX](\d+)', format_url, 'height', default=None))
229 if item_label in QUALITY_LABELS:
230 quality = quality_from_label(item_label)
231 f_id = item_label
232 elif item_label in QUALITY_FORMATS:
233 quality = quality_from_format_id(format_id)
234 f_id = format_id
235 else:
236 quality, f_id = [None] * 2
237 formats.append({
238 'url': format_url,
239 'format_id': f_id,
240 'width': width,
241 'height': height,
242 'quality': quality,
243 })
244
245 # Example: http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706
246 if item.get('contentType') in ('url', 'audio'):
247 add_format_url(item_url)
248 continue
249
250 try:
251 stream_info = self._download_json(
252 item_url + '&type=json', video_id,
253 'Downloading %s stream JSON'
254 % item_label or item.get('format') or format_id or num)
255 except ExtractorError as ee:
256 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
257 error = (self._parse_json(
258 ee.cause.read().decode(), video_id,
259 fatal=False) or {}).get('errorstring')
260 if error:
261 raise ExtractorError(error, expected=True)
262 raise
263 # Stream URL instead of JSON, example: npo:LI_NL1_4188102
264 if isinstance(stream_info, compat_str):
265 if not stream_info.startswith('http'):
266 continue
267 video_url = stream_info
268 # JSON
269 else:
270 video_url = stream_info.get('url')
271 if not video_url or video_url in urls:
272 continue
273 urls.add(item_url)
274 if determine_ext(video_url) == 'm3u8':
275 formats.extend(self._extract_m3u8_formats(
276 video_url, video_id, ext='mp4',
277 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
278 else:
279 add_format_url(video_url)
280
281 is_live = metadata.get('medium') == 'live'
282
283 if not is_live:
284 for num, stream in enumerate(metadata.get('streams', [])):
285 stream_url = stream.get('url')
286 if not is_legal_url(stream_url):
287 continue
288 urls.add(stream_url)
289 # smooth streaming is not supported
290 stream_type = stream.get('type', '').lower()
291 if stream_type in ['ss', 'ms']:
292 continue
293 if stream_type == 'hds':
294 f4m_formats = self._extract_f4m_formats(
295 stream_url, video_id, fatal=False)
296 # f4m downloader downloads only piece of live stream
297 for f4m_format in f4m_formats:
298 f4m_format['preference'] = -1
299 formats.extend(f4m_formats)
300 elif stream_type == 'hls':
301 formats.extend(self._extract_m3u8_formats(
302 stream_url, video_id, ext='mp4', fatal=False))
303 # Example: http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706
304 elif '.asf' in stream_url:
305 asx = self._download_xml(
306 stream_url, video_id,
307 'Downloading stream %d ASX playlist' % num,
308 transform_source=fix_xml_ampersands, fatal=False)
309 if not asx:
310 continue
311 ref = asx.find('./ENTRY/Ref')
312 if ref is None:
313 continue
314 video_url = ref.get('href')
315 if not video_url or video_url in urls:
316 continue
317 urls.add(video_url)
318 formats.append({
319 'url': video_url,
320 'ext': stream.get('formaat', 'asf'),
321 'quality': stream.get('kwaliteit'),
322 'preference': -10,
323 })
324 else:
325 formats.append({
326 'url': stream_url,
327 'quality': stream.get('kwaliteit'),
328 })
329
330 self._sort_formats(formats)
331
332 subtitles = {}
333 if metadata.get('tt888') == 'ja':
334 subtitles['nl'] = [{
335 'ext': 'vtt',
336 'url': 'http://tt888.omroep.nl/tt888/%s' % video_id,
337 }]
338
339 return {
340 'id': video_id,
341 'title': self._live_title(title) if is_live else title,
342 'description': metadata.get('info'),
343 'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
344 'upload_date': unified_strdate(metadata.get('gidsdatum')),
345 'duration': parse_duration(metadata.get('tijdsduur')),
346 'formats': formats,
347 'subtitles': subtitles,
348 'is_live': is_live,
349 }
350
351
352 class NPOLiveIE(NPOBaseIE):
353 IE_NAME = 'npo.nl:live'
354 _VALID_URL = r'https?://(?:www\.)?npo\.nl/live(?:/(?P<id>[^/?#&]+))?'
355
356 _TESTS = [{
357 'url': 'http://www.npo.nl/live/npo-1',
358 'info_dict': {
359 'id': 'LI_NL1_4188102',
360 'display_id': 'npo-1',
361 'ext': 'mp4',
362 'title': 're:^NPO 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
363 'is_live': True,
364 },
365 'params': {
366 'skip_download': True,
367 }
368 }, {
369 'url': 'http://www.npo.nl/live',
370 'only_matching': True,
371 }]
372
373 def _real_extract(self, url):
374 display_id = self._match_id(url) or 'npo-1'
375
376 webpage = self._download_webpage(url, display_id)
377
378 live_id = self._search_regex(
379 [r'media-id="([^"]+)"', r'data-prid="([^"]+)"'], webpage, 'live id')
380
381 return {
382 '_type': 'url_transparent',
383 'url': 'npo:%s' % live_id,
384 'ie_key': NPOIE.ie_key(),
385 'id': live_id,
386 'display_id': display_id,
387 }
388
389
390 class NPORadioIE(InfoExtractor):
391 IE_NAME = 'npo.nl:radio'
392 _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
393
394 _TEST = {
395 'url': 'http://www.npo.nl/radio/radio-1',
396 'info_dict': {
397 'id': 'radio-1',
398 'ext': 'mp3',
399 'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
400 'is_live': True,
401 },
402 'params': {
403 'skip_download': True,
404 }
405 }
406
407 @staticmethod
408 def _html_get_attribute_regex(attribute):
409 return r'{0}\s*=\s*\'([^\']+)\''.format(attribute)
410
411 def _real_extract(self, url):
412 video_id = self._match_id(url)
413
414 webpage = self._download_webpage(url, video_id)
415
416 title = self._html_search_regex(
417 self._html_get_attribute_regex('data-channel'), webpage, 'title')
418
419 stream = self._parse_json(
420 self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
421 video_id)
422
423 codec = stream.get('codec')
424
425 return {
426 'id': video_id,
427 'url': stream['url'],
428 'title': self._live_title(title),
429 'acodec': codec,
430 'ext': codec,
431 'is_live': True,
432 }
433
434
435 class NPORadioFragmentIE(InfoExtractor):
436 IE_NAME = 'npo.nl:radio:fragment'
437 _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
438
439 _TEST = {
440 'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
441 'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
442 'info_dict': {
443 'id': '174356',
444 'ext': 'mp3',
445 'title': 'Jubileumconcert Willeke Alberti',
446 },
447 }
448
449 def _real_extract(self, url):
450 audio_id = self._match_id(url)
451
452 webpage = self._download_webpage(url, audio_id)
453
454 title = self._html_search_regex(
455 r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
456 webpage, 'title')
457
458 audio_url = self._search_regex(
459 r"data-streams='([^']+)'", webpage, 'audio url')
460
461 return {
462 'id': audio_id,
463 'url': audio_url,
464 'title': title,
465 }
466
467
468 class NPODataMidEmbedIE(InfoExtractor):
469 def _real_extract(self, url):
470 display_id = self._match_id(url)
471 webpage = self._download_webpage(url, display_id)
472 video_id = self._search_regex(
473 r'data-mid=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video_id', group='id')
474 return {
475 '_type': 'url_transparent',
476 'ie_key': 'NPO',
477 'url': 'npo:%s' % video_id,
478 'display_id': display_id
479 }
480
481
482 class SchoolTVIE(NPODataMidEmbedIE):
483 IE_NAME = 'schooltv'
484 _VALID_URL = r'https?://(?:www\.)?schooltv\.nl/video/(?P<id>[^/?#&]+)'
485
486 _TEST = {
487 'url': 'http://www.schooltv.nl/video/ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam/',
488 'info_dict': {
489 'id': 'WO_NTR_429477',
490 'display_id': 'ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam',
491 'title': 'Ademhaling: De hele dag haal je adem. Maar wat gebeurt er dan eigenlijk in je lichaam?',
492 'ext': 'mp4',
493 'description': 'md5:abfa0ff690adb73fd0297fd033aaa631'
494 },
495 'params': {
496 # Skip because of m3u8 download
497 'skip_download': True
498 }
499 }
500
501
502 class HetKlokhuisIE(NPODataMidEmbedIE):
503 IE_NAME = 'hetklokhuis'
504 _VALID_URL = r'https?://(?:www\.)?hetklokhuis\.nl/[^/]+/\d+/(?P<id>[^/?#&]+)'
505
506 _TEST = {
507 'url': 'http://hetklokhuis.nl/tv-uitzending/3471/Zwaartekrachtsgolven',
508 'info_dict': {
509 'id': 'VPWON_1260528',
510 'display_id': 'Zwaartekrachtsgolven',
511 'ext': 'm4v',
512 'title': 'Het Klokhuis: Zwaartekrachtsgolven',
513 'description': 'md5:c94f31fb930d76c2efa4a4a71651dd48',
514 'upload_date': '20170223',
515 },
516 'params': {
517 'skip_download': True
518 }
519 }
520
521
522 class NPOPlaylistBaseIE(NPOIE):
523 def _real_extract(self, url):
524 playlist_id = self._match_id(url)
525
526 webpage = self._download_webpage(url, playlist_id)
527
528 entries = [
529 self.url_result('npo:%s' % video_id if not video_id.startswith('http') else video_id)
530 for video_id in orderedSet(re.findall(self._PLAYLIST_ENTRY_RE, webpage))
531 ]
532
533 playlist_title = self._html_search_regex(
534 self._PLAYLIST_TITLE_RE, webpage, 'playlist title',
535 default=None) or self._og_search_title(webpage)
536
537 return self.playlist_result(entries, playlist_id, playlist_title)
538
539
540 class VPROIE(NPOPlaylistBaseIE):
541 IE_NAME = 'vpro'
542 _VALID_URL = r'https?://(?:www\.)?(?:(?:tegenlicht\.)?vpro|2doc)\.nl/(?:[^/]+/)*(?P<id>[^/]+)\.html'
543 _PLAYLIST_TITLE_RE = (r'<h1[^>]+class=["\'].*?\bmedia-platform-title\b.*?["\'][^>]*>([^<]+)',
544 r'<h5[^>]+class=["\'].*?\bmedia-platform-subtitle\b.*?["\'][^>]*>([^<]+)')
545 _PLAYLIST_ENTRY_RE = r'data-media-id="([^"]+)"'
546
547 _TESTS = [
548 {
549 'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
550 'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
551 'info_dict': {
552 'id': 'VPWON_1169289',
553 'ext': 'm4v',
554 'title': 'De toekomst komt uit Afrika',
555 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
556 'upload_date': '20130225',
557 },
558 'skip': 'Video gone',
559 },
560 {
561 'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html',
562 'info_dict': {
563 'id': 'sergio-herman',
564 'title': 'sergio herman: fucking perfect',
565 },
566 'playlist_count': 2,
567 },
568 {
569 # playlist with youtube embed
570 'url': 'http://www.vpro.nl/programmas/2doc/2015/education-education.html',
571 'info_dict': {
572 'id': 'education-education',
573 'title': 'education education',
574 },
575 'playlist_count': 2,
576 },
577 {
578 'url': 'http://www.2doc.nl/documentaires/series/2doc/2015/oktober/de-tegenprestatie.html',
579 'info_dict': {
580 'id': 'de-tegenprestatie',
581 'title': 'De Tegenprestatie',
582 },
583 'playlist_count': 2,
584 }, {
585 'url': 'http://www.2doc.nl/speel~VARA_101375237~mh17-het-verdriet-van-nederland~.html',
586 'info_dict': {
587 'id': 'VARA_101375237',
588 'ext': 'm4v',
589 'title': 'MH17: Het verdriet van Nederland',
590 'description': 'md5:09e1a37c1fdb144621e22479691a9f18',
591 'upload_date': '20150716',
592 },
593 'params': {
594 # Skip because of m3u8 download
595 'skip_download': True
596 },
597 }
598 ]
599
600
601 class WNLIE(NPOPlaylistBaseIE):
602 IE_NAME = 'wnl'
603 _VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+'
604 _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>'
605 _PLAYLIST_ENTRY_RE = r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>Deel \d+'
606
607 _TESTS = [{
608 'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515',
609 'info_dict': {
610 'id': 'vandaag-de-dag-6-mei',
611 'title': 'Vandaag de Dag 6 mei',
612 },
613 'playlist_count': 4,
614 }]
615
616
617 class AndereTijdenIE(NPOPlaylistBaseIE):
618 IE_NAME = 'anderetijden'
619 _VALID_URL = r'https?://(?:www\.)?anderetijden\.nl/programma/(?:[^/]+/)+(?P<id>[^/?#&]+)'
620 _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class=["\'].*?\bpage-title\b.*?["\'][^>]*>(.+?)</h1>'
621 _PLAYLIST_ENTRY_RE = r'<figure[^>]+class=["\']episode-container episode-page["\'][^>]+data-prid=["\'](.+?)["\']'
622
623 _TESTS = [{
624 'url': 'http://anderetijden.nl/programma/1/Andere-Tijden/aflevering/676/Duitse-soldaten-over-de-Slag-bij-Arnhem',
625 'info_dict': {
626 'id': 'Duitse-soldaten-over-de-Slag-bij-Arnhem',
627 'title': 'Duitse soldaten over de Slag bij Arnhem',
628 },
629 'playlist_count': 3,
630 }]