1 from __future__ import unicode_literals
5 from .common import InfoExtractor
28 class NPOBaseIE(InfoExtractor):
29 def _get_token(self, video_id):
30 return self._download_json(
31 'http://ida.omroep.nl/app.php/auth', video_id,
32 note='Downloading token')['token']
35 class NPOIE(NPOBaseIE):
37 IE_DESC = 'npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl'
45 (?:ntr|npostart)\.nl/(?:[^/]+/){2,}|
46 omroepwnl\.nl/video/fragment/[^/]+__|
47 (?:zapp|npo3)\.nl/(?:[^/]+/){2,}
54 'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
55 'md5': '4b3f9c429157ec4775f2c9cb7b911016',
57 'id': 'VPWON_1220719',
60 'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
61 'upload_date': '20140622',
64 'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
65 'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
67 'id': 'VARA_101191800',
69 'title': 'De Mega Mike & Mega Thomas show: The best of.',
70 'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
71 'upload_date': '20090227',
75 'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
76 'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
78 'id': 'VPWON_1169289',
80 'title': 'Tegenlicht: Zwart geld. De toekomst komt uit Afrika',
81 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
82 'upload_date': '20130225',
86 'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
88 'id': 'WO_VPRO_043706',
90 'title': 'De nieuwe mens - Deel 1',
91 'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
95 'skip_download': True,
99 'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
101 'id': 'WO_NOS_762771',
103 'title': 'Hoe gaat Europa verder na Parijs?',
106 'skip_download': True,
109 'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content',
111 'id': 'VPWON_1233944',
113 'title': 'Aap, poot, pies',
114 'description': 'md5:c9c8005d1869ae65b858e82c01a91fde',
115 'upload_date': '20150508',
119 'skip_download': True,
122 'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698',
124 'id': 'POW_00996502',
126 'title': '''"Dit is wel een 'landslide'..."''',
127 'description': 'md5:f8d66d537dfb641380226e31ca57b8e8',
128 'upload_date': '20150508',
132 'skip_download': True,
136 'url': 'http://www.npo.nl/jouw-stad-rotterdam/29-01-2017/RBX_FUNX_6683215/RBX_FUNX_7601437',
138 'id': 'RBX_FUNX_6683215',
140 'title': 'Jouw Stad Rotterdam',
141 'description': 'md5:db251505244f097717ec59fabc372d9f',
144 'skip_download': True,
147 'url': 'http://www.zapp.nl/de-bzt-show/gemist/KN_1687547',
148 'only_matching': True,
150 'url': 'http://www.zapp.nl/de-bzt-show/filmpjes/POMS_KN_7315118',
151 'only_matching': True,
153 'url': 'http://www.zapp.nl/beste-vrienden-quiz/extra-video-s/WO_NTR_1067990',
154 'only_matching': True,
156 'url': 'https://www.npo3.nl/3onderzoekt/16-09-2015/VPWON_1239870',
157 'only_matching': True,
160 'url': 'npo:LI_NL1_4188102',
161 'only_matching': True,
163 'url': 'http://www.npo.nl/radio-gaga/13-06-2017/BNN_101383373',
164 'only_matching': True,
166 'url': 'https://www.zapp.nl/1803-skelterlab/instructie-video-s/740-instructievideo-s/POMS_AT_11736927',
167 'only_matching': True,
169 'url': 'https://www.npostart.nl/broodje-gezond-ei/28-05-2018/KN_1698996',
170 'only_matching': True,
172 'url': 'https://npo.nl/KN_1698996',
173 'only_matching': True,
177 def suitable(cls, url):
178 return (False if any(ie.suitable(url)
179 for ie in (NPOLiveIE, NPORadioIE, NPORadioFragmentIE))
180 else super(NPOIE, cls).suitable(url))
182 def _real_extract(self, url):
183 video_id = self._match_id(url)
185 return self._get_info(url, video_id)
186 except ExtractorError:
187 return self._get_old_info(video_id)
189 def _get_info(self, url, video_id):
190 token = self._download_json(
191 'https://www.npostart.nl/api/token', video_id,
192 'Downloading token', headers={
194 'X-Requested-With': 'XMLHttpRequest',
197 player = self._download_json(
198 'https://www.npostart.nl/player/%s' % video_id, video_id,
199 'Downloading player JSON', data=urlencode_postdata({
207 player_token = player['token']
211 for profile in ('hls', 'dash-widevine', 'dash-playready', 'smooth'):
212 streams = self._download_json(
213 'https://start-player.npo.nl/video/%s/streams' % video_id,
214 video_id, 'Downloading %s profile JSON' % profile, fatal=False,
218 'tokenId': player_token,
219 'streamType': 'broadcast',
223 stream = streams.get('stream')
224 if not isinstance(stream, dict):
226 stream_url = url_or_none(stream.get('src'))
227 if not stream_url or stream_url in format_urls:
229 format_urls.add(stream_url)
230 if stream.get('protection') is not None:
232 stream_type = stream.get('type')
233 stream_ext = determine_ext(stream_url)
234 if stream_type == 'application/dash+xml' or stream_ext == 'mpd':
235 formats.extend(self._extract_mpd_formats(
236 stream_url, video_id, mpd_id='dash', fatal=False))
237 elif stream_type == 'application/vnd.apple.mpegurl' or stream_ext == 'm3u8':
238 formats.extend(self._extract_m3u8_formats(
239 stream_url, video_id, ext='mp4',
240 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
241 elif re.search(r'\.isml?/Manifest', stream_url):
242 formats.extend(self._extract_ism_formats(
243 stream_url, video_id, ism_id='mss', fatal=False))
249 self._sort_formats(formats)
257 embed_url = url_or_none(player.get('embedUrl'))
259 webpage = self._download_webpage(
260 embed_url, video_id, 'Downloading embed page', fatal=False)
262 video = self._parse_json(
264 r'\bvideo\s*=\s*({.+?})\s*;', webpage, 'video',
265 default='{}'), video_id)
267 title = video.get('episodeTitle')
269 subtitles_list = video.get('subtitles')
270 if isinstance(subtitles_list, list):
271 for cc in subtitles_list:
272 cc_url = url_or_none(cc.get('src'))
275 lang = str_or_none(cc.get('language')) or 'nl'
276 subtitles.setdefault(lang, []).append({
281 'description': video.get('description'),
282 'thumbnail': url_or_none(
283 video.get('still_image_url') or video.get('orig_image_url')),
284 'duration': int_or_none(video.get('duration')),
285 'timestamp': unified_timestamp(video.get('broadcastDate')),
286 'creator': video.get('channel'),
287 'series': video.get('title'),
289 'episode_number': int_or_none(video.get('episodeNumber')),
290 'subtitles': subtitles,
295 def _get_old_info(self, video_id):
296 metadata = self._download_json(
297 'http://e.omroep.nl/metadata/%s' % video_id,
299 # We have to remove the javascript callback
300 transform_source=strip_jsonp,
303 error = metadata.get('error')
305 raise ExtractorError(error, expected=True)
307 # For some videos actual video id (prid) is different (e.g. for
308 # http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698
309 # video id is POMS_WNL_853698 but prid is POW_00996502)
310 video_id = metadata.get('prid') or video_id
312 # titel is too generic in some cases so utilize aflevering_titel as well
313 # when available (e.g. http://tegenlicht.vpro.nl/afleveringen/2014-2015/access-to-africa.html)
314 title = metadata['titel']
315 sub_title = metadata.get('aflevering_titel')
316 if sub_title and sub_title != title:
317 title += ': %s' % sub_title
319 token = self._get_token(video_id)
324 def is_legal_url(format_url):
325 return format_url and format_url not in urls and re.match(
326 r'^(?:https?:)?//', format_url)
328 QUALITY_LABELS = ('Laag', 'Normaal', 'Hoog')
329 QUALITY_FORMATS = ('adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std')
331 quality_from_label = qualities(QUALITY_LABELS)
332 quality_from_format_id = qualities(QUALITY_FORMATS)
333 items = self._download_json(
334 'http://ida.omroep.nl/app.php/%s' % video_id, video_id,
335 'Downloading formats JSON', query={
339 for num, item in enumerate(items):
340 item_url = item.get('url')
341 if not is_legal_url(item_url):
344 format_id = self._search_regex(
345 r'video/ida/([^/]+)', item_url, 'format id',
348 item_label = item.get('label')
350 def add_format_url(format_url):
351 width = int_or_none(self._search_regex(
352 r'(\d+)[xX]\d+', format_url, 'width', default=None))
353 height = int_or_none(self._search_regex(
354 r'\d+[xX](\d+)', format_url, 'height', default=None))
355 if item_label in QUALITY_LABELS:
356 quality = quality_from_label(item_label)
358 elif item_label in QUALITY_FORMATS:
359 quality = quality_from_format_id(format_id)
362 quality, f_id = [None] * 2
371 # Example: http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706
372 if item.get('contentType') in ('url', 'audio'):
373 add_format_url(item_url)
377 stream_info = self._download_json(
378 item_url + '&type=json', video_id,
379 'Downloading %s stream JSON'
380 % item_label or item.get('format') or format_id or num)
381 except ExtractorError as ee:
382 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
383 error = (self._parse_json(
384 ee.cause.read().decode(), video_id,
385 fatal=False) or {}).get('errorstring')
387 raise ExtractorError(error, expected=True)
389 # Stream URL instead of JSON, example: npo:LI_NL1_4188102
390 if isinstance(stream_info, compat_str):
391 if not stream_info.startswith('http'):
393 video_url = stream_info
396 video_url = stream_info.get('url')
397 if not video_url or 'vodnotavailable.' in video_url or video_url in urls:
400 if determine_ext(video_url) == 'm3u8':
401 formats.extend(self._extract_m3u8_formats(
402 video_url, video_id, ext='mp4',
403 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
405 add_format_url(video_url)
407 is_live = metadata.get('medium') == 'live'
410 for num, stream in enumerate(metadata.get('streams', [])):
411 stream_url = stream.get('url')
412 if not is_legal_url(stream_url):
415 # smooth streaming is not supported
416 stream_type = stream.get('type', '').lower()
417 if stream_type in ['ss', 'ms']:
419 if stream_type == 'hds':
420 f4m_formats = self._extract_f4m_formats(
421 stream_url, video_id, fatal=False)
422 # f4m downloader downloads only piece of live stream
423 for f4m_format in f4m_formats:
424 f4m_format['preference'] = -1
425 formats.extend(f4m_formats)
426 elif stream_type == 'hls':
427 formats.extend(self._extract_m3u8_formats(
428 stream_url, video_id, ext='mp4', fatal=False))
429 # Example: http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706
430 elif '.asf' in stream_url:
431 asx = self._download_xml(
432 stream_url, video_id,
433 'Downloading stream %d ASX playlist' % num,
434 transform_source=fix_xml_ampersands, fatal=False)
437 ref = asx.find('./ENTRY/Ref')
440 video_url = ref.get('href')
441 if not video_url or video_url in urls:
446 'ext': stream.get('formaat', 'asf'),
447 'quality': stream.get('kwaliteit'),
453 'quality': stream.get('kwaliteit'),
456 self._sort_formats(formats)
459 if metadata.get('tt888') == 'ja':
462 'url': 'http://tt888.omroep.nl/tt888/%s' % video_id,
467 'title': self._live_title(title) if is_live else title,
468 'description': metadata.get('info'),
469 'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
470 'upload_date': unified_strdate(metadata.get('gidsdatum')),
471 'duration': parse_duration(metadata.get('tijdsduur')),
473 'subtitles': subtitles,
478 class NPOLiveIE(NPOBaseIE):
479 IE_NAME = 'npo.nl:live'
480 _VALID_URL = r'https?://(?:www\.)?npo(?:start)?\.nl/live(?:/(?P<id>[^/?#&]+))?'
483 'url': 'http://www.npo.nl/live/npo-1',
485 'id': 'LI_NL1_4188102',
486 'display_id': 'npo-1',
488 'title': 're:^NPO 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
492 'skip_download': True,
495 'url': 'http://www.npo.nl/live',
496 'only_matching': True,
498 'url': 'https://www.npostart.nl/live/npo-1',
499 'only_matching': True,
502 def _real_extract(self, url):
503 display_id = self._match_id(url) or 'npo-1'
505 webpage = self._download_webpage(url, display_id)
507 live_id = self._search_regex(
508 [r'media-id="([^"]+)"', r'data-prid="([^"]+)"'], webpage, 'live id')
511 '_type': 'url_transparent',
512 'url': 'npo:%s' % live_id,
513 'ie_key': NPOIE.ie_key(),
515 'display_id': display_id,
519 class NPORadioIE(InfoExtractor):
520 IE_NAME = 'npo.nl:radio'
521 _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)'
524 'url': 'http://www.npo.nl/radio/radio-1',
528 'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
532 'skip_download': True,
537 def suitable(cls, url):
538 return False if NPORadioFragmentIE.suitable(url) else super(NPORadioIE, cls).suitable(url)
541 def _html_get_attribute_regex(attribute):
542 return r'{0}\s*=\s*\'([^\']+)\''.format(attribute)
544 def _real_extract(self, url):
545 video_id = self._match_id(url)
547 webpage = self._download_webpage(url, video_id)
549 title = self._html_search_regex(
550 self._html_get_attribute_regex('data-channel'), webpage, 'title')
552 stream = self._parse_json(
553 self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
556 codec = stream.get('codec')
560 'url': stream['url'],
561 'title': self._live_title(title),
568 class NPORadioFragmentIE(InfoExtractor):
569 IE_NAME = 'npo.nl:radio:fragment'
570 _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
573 'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
574 'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
578 'title': 'Jubileumconcert Willeke Alberti',
582 def _real_extract(self, url):
583 audio_id = self._match_id(url)
585 webpage = self._download_webpage(url, audio_id)
587 title = self._html_search_regex(
588 r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
591 audio_url = self._search_regex(
592 r"data-streams='([^']+)'", webpage, 'audio url')
601 class NPODataMidEmbedIE(InfoExtractor):
602 def _real_extract(self, url):
603 display_id = self._match_id(url)
604 webpage = self._download_webpage(url, display_id)
605 video_id = self._search_regex(
606 r'data-mid=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video_id', group='id')
608 '_type': 'url_transparent',
610 'url': 'npo:%s' % video_id,
611 'display_id': display_id
615 class SchoolTVIE(NPODataMidEmbedIE):
617 _VALID_URL = r'https?://(?:www\.)?schooltv\.nl/video/(?P<id>[^/?#&]+)'
620 'url': 'http://www.schooltv.nl/video/ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam/',
622 'id': 'WO_NTR_429477',
623 'display_id': 'ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam',
624 'title': 'Ademhaling: De hele dag haal je adem. Maar wat gebeurt er dan eigenlijk in je lichaam?',
626 'description': 'md5:abfa0ff690adb73fd0297fd033aaa631'
629 # Skip because of m3u8 download
630 'skip_download': True
635 class HetKlokhuisIE(NPODataMidEmbedIE):
636 IE_NAME = 'hetklokhuis'
637 _VALID_URL = r'https?://(?:www\.)?hetklokhuis\.nl/[^/]+/\d+/(?P<id>[^/?#&]+)'
640 'url': 'http://hetklokhuis.nl/tv-uitzending/3471/Zwaartekrachtsgolven',
642 'id': 'VPWON_1260528',
643 'display_id': 'Zwaartekrachtsgolven',
645 'title': 'Het Klokhuis: Zwaartekrachtsgolven',
646 'description': 'md5:c94f31fb930d76c2efa4a4a71651dd48',
647 'upload_date': '20170223',
650 'skip_download': True
655 class NPOPlaylistBaseIE(NPOIE):
656 def _real_extract(self, url):
657 playlist_id = self._match_id(url)
659 webpage = self._download_webpage(url, playlist_id)
662 self.url_result('npo:%s' % video_id if not video_id.startswith('http') else video_id)
663 for video_id in orderedSet(re.findall(self._PLAYLIST_ENTRY_RE, webpage))
666 playlist_title = self._html_search_regex(
667 self._PLAYLIST_TITLE_RE, webpage, 'playlist title',
668 default=None) or self._og_search_title(webpage)
670 return self.playlist_result(entries, playlist_id, playlist_title)
673 class VPROIE(NPOPlaylistBaseIE):
675 _VALID_URL = r'https?://(?:www\.)?(?:(?:tegenlicht\.)?vpro|2doc)\.nl/(?:[^/]+/)*(?P<id>[^/]+)\.html'
676 _PLAYLIST_TITLE_RE = (r'<h1[^>]+class=["\'].*?\bmedia-platform-title\b.*?["\'][^>]*>([^<]+)',
677 r'<h5[^>]+class=["\'].*?\bmedia-platform-subtitle\b.*?["\'][^>]*>([^<]+)')
678 _PLAYLIST_ENTRY_RE = r'data-media-id="([^"]+)"'
682 'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
683 'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
685 'id': 'VPWON_1169289',
687 'title': 'De toekomst komt uit Afrika',
688 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
689 'upload_date': '20130225',
691 'skip': 'Video gone',
694 'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html',
696 'id': 'sergio-herman',
697 'title': 'sergio herman: fucking perfect',
702 # playlist with youtube embed
703 'url': 'http://www.vpro.nl/programmas/2doc/2015/education-education.html',
705 'id': 'education-education',
706 'title': 'education education',
711 'url': 'http://www.2doc.nl/documentaires/series/2doc/2015/oktober/de-tegenprestatie.html',
713 'id': 'de-tegenprestatie',
714 'title': 'De Tegenprestatie',
718 'url': 'http://www.2doc.nl/speel~VARA_101375237~mh17-het-verdriet-van-nederland~.html',
720 'id': 'VARA_101375237',
722 'title': 'MH17: Het verdriet van Nederland',
723 'description': 'md5:09e1a37c1fdb144621e22479691a9f18',
724 'upload_date': '20150716',
727 # Skip because of m3u8 download
728 'skip_download': True
734 class WNLIE(NPOPlaylistBaseIE):
736 _VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+'
737 _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>'
738 _PLAYLIST_ENTRY_RE = r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>Deel \d+'
741 'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515',
743 'id': 'vandaag-de-dag-6-mei',
744 'title': 'Vandaag de Dag 6 mei',
750 class AndereTijdenIE(NPOPlaylistBaseIE):
751 IE_NAME = 'anderetijden'
752 _VALID_URL = r'https?://(?:www\.)?anderetijden\.nl/programma/(?:[^/]+/)+(?P<id>[^/?#&]+)'
753 _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class=["\'].*?\bpage-title\b.*?["\'][^>]*>(.+?)</h1>'
754 _PLAYLIST_ENTRY_RE = r'<figure[^>]+class=["\']episode-container episode-page["\'][^>]+data-prid=["\'](.+?)["\']'
757 'url': 'http://anderetijden.nl/programma/1/Andere-Tijden/aflevering/676/Duitse-soldaten-over-de-Slag-bij-Arnhem',
759 'id': 'Duitse-soldaten-over-de-Slag-bij-Arnhem',
760 'title': 'Duitse soldaten over de Slag bij Arnhem',