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 compat_HTTPError
   7 from ..utils import (
   8     fix_xml_ampersands,
   9     orderedSet,
  10     parse_duration,
  11     qualities,
  12     strip_jsonp,
  13     unified_strdate,
  14     ExtractorError,
  15 )
  16
  17
  18 class NPOBaseIE(InfoExtractor):
  19     def _get_token(self, video_id):
  20         token_page = self._download_webpage(
  21             'http://ida.omroep.nl/npoplayer/i.js',
  22             video_id, note='Downloading token')
  23         token = self._search_regex(
  24             r'npoplayer\.token = "(.+?)"', token_page, 'token')
  25         # Decryption algorithm extracted from http://npoplayer.omroep.nl/csjs/npoplayer-min.js
  26         token_l = list(token)
  27         first = second = None
  28         for i in range(5, len(token_l) - 4):
  29             if token_l[i].isdigit():
  30                 if first is None:
  31                     first = i
  32                 elif second is None:
  33                     second = i
  34         if first is None or second is None:
  35             first = 12
  36             second = 13
  37
  38         token_l[first], token_l[second] = token_l[second], token_l[first]
  39
  40         return ''.join(token_l)
  41
  42
  43 class NPOIE(NPOBaseIE):
  44     IE_NAME = 'npo'
  45     IE_DESC = 'npo.nl and ntr.nl'
  46     _VALID_URL = r'''(?x)
  47                     (?:
  48                         npo:|
  49                         https?://
  50                             (?:www\.)?
  51                             (?:
  52                                 npo\.nl/(?!live|radio)(?:[^/]+/){2}|
  53                                 ntr\.nl/(?:[^/]+/){2,}|
  54                                 omroepwnl\.nl/video/fragment/[^/]+__|
  55                                 zapp\.nl/[^/]+/(?:gemist|filmpjes)/
  56                             )
  57                         )
  58                         (?P<id>[^/?#]+)
  59                 '''
  60
  61     _TESTS = [
  62         {
  63             'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
  64             'md5': '4b3f9c429157ec4775f2c9cb7b911016',
  65             'info_dict': {
  66                 'id': 'VPWON_1220719',
  67                 'ext': 'm4v',
  68                 'title': 'Nieuwsuur',
  69                 'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
  70                 'upload_date': '20140622',
  71             },
  72         },
  73         {
  74             'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
  75             'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
  76             'info_dict': {
  77                 'id': 'VARA_101191800',
  78                 'ext': 'm4v',
  79                 'title': 'De Mega Mike & Mega Thomas show: The best of.',
  80                 'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
  81                 'upload_date': '20090227',
  82                 'duration': 2400,
  83             },
  84         },
  85         {
  86             'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
  87             'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
  88             'info_dict': {
  89                 'id': 'VPWON_1169289',
  90                 'ext': 'm4v',
  91                 'title': 'Tegenlicht: De toekomst komt uit Afrika',
  92                 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
  93                 'upload_date': '20130225',
  94                 'duration': 3000,
  95             },
  96         },
  97         {
  98             'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
  99             'info_dict': {
 100                 'id': 'WO_VPRO_043706',
 101                 'ext': 'wmv',
 102                 'title': 'De nieuwe mens - Deel 1',
 103                 'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
 104                 'duration': 4680,
 105             },
 106             'params': {
 107                 # mplayer mms download
 108                 'skip_download': True,
 109             }
 110         },
 111         # non asf in streams
 112         {
 113             'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
 114             'md5': 'b3da13de374cbe2d5332a7e910bef97f',
 115             'info_dict': {
 116                 'id': 'WO_NOS_762771',
 117                 'ext': 'mp4',
 118                 'title': 'Hoe gaat Europa verder na Parijs?',
 119             },
 120         },
 121         {
 122             'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content',
 123             'md5': '01c6a2841675995da1f0cf776f03a9c3',
 124             'info_dict': {
 125                 'id': 'VPWON_1233944',
 126                 'ext': 'm4v',
 127                 'title': 'Aap, poot, pies',
 128                 'description': 'md5:c9c8005d1869ae65b858e82c01a91fde',
 129                 'upload_date': '20150508',
 130                 'duration': 599,
 131             },
 132         },
 133         {
 134             'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698',
 135             'md5': 'd30cd8417b8b9bca1fdff27428860d08',
 136             'info_dict': {
 137                 'id': 'POW_00996502',
 138                 'ext': 'm4v',
 139                 'title': '''"Dit is wel een 'landslide'..."''',
 140                 'description': 'md5:f8d66d537dfb641380226e31ca57b8e8',
 141                 'upload_date': '20150508',
 142                 'duration': 462,
 143             },
 144         },
 145         {
 146             'url': 'http://www.zapp.nl/de-bzt-show/gemist/KN_1687547',
 147             'only_matching': True,
 148         },
 149         {
 150             'url': 'http://www.zapp.nl/de-bzt-show/filmpjes/POMS_KN_7315118',
 151             'only_matching': True,
 152         }
 153     ]
 154
 155     def _real_extract(self, url):
 156         video_id = self._match_id(url)
 157         return self._get_info(video_id)
 158
 159     def _get_info(self, video_id):
 160         metadata = self._download_json(
 161             'http://e.omroep.nl/metadata/%s' % video_id,
 162             video_id,
 163             # We have to remove the javascript callback
 164             transform_source=strip_jsonp,
 165         )
 166
 167         # For some videos actual video id (prid) is different (e.g. for
 168         # http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698
 169         # video id is POMS_WNL_853698 but prid is POW_00996502)
 170         video_id = metadata.get('prid') or video_id
 171
 172         # titel is too generic in some cases so utilize aflevering_titel as well
 173         # when available (e.g. http://tegenlicht.vpro.nl/afleveringen/2014-2015/access-to-africa.html)
 174         title = metadata['titel']
 175         sub_title = metadata.get('aflevering_titel')
 176         if sub_title and sub_title != title:
 177             title += ': %s' % sub_title
 178
 179         token = self._get_token(video_id)
 180
 181         formats = []
 182
 183         pubopties = metadata.get('pubopties')
 184         if pubopties:
 185             quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std'])
 186             for format_id in pubopties:
 187                 format_info = self._download_json(
 188                     'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s'
 189                     % (video_id, format_id, token),
 190                     video_id, 'Downloading %s JSON' % format_id)
 191                 if format_info.get('error_code', 0) or format_info.get('errorcode', 0):
 192                     continue
 193                 streams = format_info.get('streams')
 194                 if streams:
 195                     try:
 196                         video_info = self._download_json(
 197                             streams[0] + '&type=json',
 198                             video_id, 'Downloading %s stream JSON' % format_id)
 199                     except ExtractorError as ee:
 200                         if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
 201                             error = (self._parse_json(ee.cause.read().decode(), video_id, fatal=False) or {}).get('errorstring')
 202                             if error:
 203                                 raise ExtractorError(error, expected=True)
 204                         raise
 205                 else:
 206                     video_info = format_info
 207                 video_url = video_info.get('url')
 208                 if not video_url:
 209                     continue
 210                 if format_id == 'adaptive':
 211                     formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4'))
 212                 else:
 213                     formats.append({
 214                         'url': video_url,
 215                         'format_id': format_id,
 216                         'quality': quality(format_id),
 217                     })
 218
 219         streams = metadata.get('streams')
 220         if streams:
 221             for i, stream in enumerate(streams):
 222                 stream_url = stream.get('url')
 223                 if not stream_url:
 224                     continue
 225                 if '.asf' not in stream_url:
 226                     formats.append({
 227                         'url': stream_url,
 228                         'quality': stream.get('kwaliteit'),
 229                     })
 230                     continue
 231                 asx = self._download_xml(
 232                     stream_url, video_id,
 233                     'Downloading stream %d ASX playlist' % i,
 234                     transform_source=fix_xml_ampersands)
 235                 ref = asx.find('./ENTRY/Ref')
 236                 if ref is None:
 237                     continue
 238                 video_url = ref.get('href')
 239                 if not video_url:
 240                     continue
 241                 formats.append({
 242                     'url': video_url,
 243                     'ext': stream.get('formaat', 'asf'),
 244                     'quality': stream.get('kwaliteit'),
 245                 })
 246
 247         self._sort_formats(formats)
 248
 249         subtitles = {}
 250         if metadata.get('tt888') == 'ja':
 251             subtitles['nl'] = [{
 252                 'ext': 'vtt',
 253                 'url': 'http://tt888.omroep.nl/tt888/%s' % video_id,
 254             }]
 255
 256         return {
 257             'id': video_id,
 258             'title': title,
 259             'description': metadata.get('info'),
 260             'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
 261             'upload_date': unified_strdate(metadata.get('gidsdatum')),
 262             'duration': parse_duration(metadata.get('tijdsduur')),
 263             'formats': formats,
 264             'subtitles': subtitles,
 265         }
 266
 267
 268 class NPOLiveIE(NPOBaseIE):
 269     IE_NAME = 'npo.nl:live'
 270     _VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P<id>.+)'
 271
 272     _TEST = {
 273         'url': 'http://www.npo.nl/live/npo-1',
 274         'info_dict': {
 275             'id': 'LI_NEDERLAND1_136692',
 276             'display_id': 'npo-1',
 277             'ext': 'mp4',
 278             'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 279             'description': 'Livestream',
 280             'is_live': True,
 281         },
 282         'params': {
 283             'skip_download': True,
 284         }
 285     }
 286
 287     def _real_extract(self, url):
 288         display_id = self._match_id(url)
 289
 290         webpage = self._download_webpage(url, display_id)
 291
 292         live_id = self._search_regex(
 293             r'data-prid="([^"]+)"', webpage, 'live id')
 294
 295         metadata = self._download_json(
 296             'http://e.omroep.nl/metadata/%s' % live_id,
 297             display_id, transform_source=strip_jsonp)
 298
 299         token = self._get_token(display_id)
 300
 301         formats = []
 302
 303         streams = metadata.get('streams')
 304         if streams:
 305             for stream in streams:
 306                 stream_type = stream.get('type').lower()
 307                 # smooth streaming is not supported
 308                 if stream_type in ['ss', 'ms']:
 309                     continue
 310                 stream_info = self._download_json(
 311                     'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp'
 312                     % (stream.get('url'), token),
 313                     display_id, 'Downloading %s JSON' % stream_type)
 314                 if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0):
 315                     continue
 316                 stream_url = self._download_json(
 317                     stream_info['stream'], display_id,
 318                     'Downloading %s URL' % stream_type,
 319                     'Unable to download %s URL' % stream_type,
 320                     transform_source=strip_jsonp, fatal=False)
 321                 if not stream_url:
 322                     continue
 323                 if stream_type == 'hds':
 324                     f4m_formats = self._extract_f4m_formats(stream_url, display_id)
 325                     # f4m downloader downloads only piece of live stream
 326                     for f4m_format in f4m_formats:
 327                         f4m_format['preference'] = -1
 328                     formats.extend(f4m_formats)
 329                 elif stream_type == 'hls':
 330                     formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4'))
 331                 else:
 332                     formats.append({
 333                         'url': stream_url,
 334                         'preference': -10,
 335                     })
 336
 337         self._sort_formats(formats)
 338
 339         return {
 340             'id': live_id,
 341             'display_id': display_id,
 342             'title': self._live_title(metadata['titel']),
 343             'description': metadata['info'],
 344             'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
 345             'formats': formats,
 346             'is_live': True,
 347         }
 348
 349
 350 class NPORadioIE(InfoExtractor):
 351     IE_NAME = 'npo.nl:radio'
 352     _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
 353
 354     _TEST = {
 355         'url': 'http://www.npo.nl/radio/radio-1',
 356         'info_dict': {
 357             'id': 'radio-1',
 358             'ext': 'mp3',
 359             'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 360             'is_live': True,
 361         },
 362         'params': {
 363             'skip_download': True,
 364         }
 365     }
 366
 367     @staticmethod
 368     def _html_get_attribute_regex(attribute):
 369         return r'{0}\s*=\s*\'([^\']+)\''.format(attribute)
 370
 371     def _real_extract(self, url):
 372         video_id = self._match_id(url)
 373
 374         webpage = self._download_webpage(url, video_id)
 375
 376         title = self._html_search_regex(
 377             self._html_get_attribute_regex('data-channel'), webpage, 'title')
 378
 379         stream = self._parse_json(
 380             self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
 381             video_id)
 382
 383         codec = stream.get('codec')
 384
 385         return {
 386             'id': video_id,
 387             'url': stream['url'],
 388             'title': self._live_title(title),
 389             'acodec': codec,
 390             'ext': codec,
 391             'is_live': True,
 392         }
 393
 394
 395 class NPORadioFragmentIE(InfoExtractor):
 396     IE_NAME = 'npo.nl:radio:fragment'
 397     _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
 398
 399     _TEST = {
 400         'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
 401         'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
 402         'info_dict': {
 403             'id': '174356',
 404             'ext': 'mp3',
 405             'title': 'Jubileumconcert Willeke Alberti',
 406         },
 407     }
 408
 409     def _real_extract(self, url):
 410         audio_id = self._match_id(url)
 411
 412         webpage = self._download_webpage(url, audio_id)
 413
 414         title = self._html_search_regex(
 415             r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
 416             webpage, 'title')
 417
 418         audio_url = self._search_regex(
 419             r"data-streams='([^']+)'", webpage, 'audio url')
 420
 421         return {
 422             'id': audio_id,
 423             'url': audio_url,
 424             'title': title,
 425         }
 426
 427
 428 class NPODataMidEmbedIE(InfoExtractor):
 429     def _real_extract(self, url):
 430         display_id = self._match_id(url)
 431         webpage = self._download_webpage(url, display_id)
 432         video_id = self._search_regex(
 433             r'data-mid=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video_id', group='id')
 434         return {
 435             '_type': 'url_transparent',
 436             'ie_key': 'NPO',
 437             'url': 'npo:%s' % video_id,
 438             'display_id': display_id
 439         }
 440
 441
 442 class SchoolTVIE(NPODataMidEmbedIE):
 443     IE_NAME = 'schooltv'
 444     _VALID_URL = r'https?://(?:www\.)?schooltv\.nl/video/(?P<id>[^/?#&]+)'
 445
 446     _TEST = {
 447         'url': 'http://www.schooltv.nl/video/ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam/',
 448         'info_dict': {
 449             'id': 'WO_NTR_429477',
 450             'display_id': 'ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam',
 451             'title': 'Ademhaling: De hele dag haal je adem. Maar wat gebeurt er dan eigenlijk in je lichaam?',
 452             'ext': 'mp4',
 453             'description': 'md5:abfa0ff690adb73fd0297fd033aaa631'
 454         },
 455         'params': {
 456             # Skip because of m3u8 download
 457             'skip_download': True
 458         }
 459     }
 460
 461
 462 class HetKlokhuisIE(NPODataMidEmbedIE):
 463     IE_NAME = 'hetklokhuis'
 464     _VALID_URL = r'https?://(?:www\.)?hetklokhuis.nl/[^/]+/\d+/(?P<id>[^/?#&]+)'
 465
 466     _TEST = {
 467         'url': 'http://hetklokhuis.nl/tv-uitzending/3471/Zwaartekrachtsgolven',
 468         'info_dict': {
 469             'id': 'VPWON_1260528',
 470             'display_id': 'Zwaartekrachtsgolven',
 471             'ext': 'm4v',
 472             'title': 'Het Klokhuis: Zwaartekrachtsgolven',
 473             'description': 'md5:c94f31fb930d76c2efa4a4a71651dd48',
 474             'upload_date': '20170223',
 475         },
 476         'params': {
 477             'skip_download': True
 478         }
 479     }
 480
 481
 482 class NPOPlaylistBaseIE(NPOIE):
 483     def _real_extract(self, url):
 484         playlist_id = self._match_id(url)
 485
 486         webpage = self._download_webpage(url, playlist_id)
 487
 488         entries = [
 489             self.url_result('npo:%s' % video_id if not video_id.startswith('http') else video_id)
 490             for video_id in orderedSet(re.findall(self._PLAYLIST_ENTRY_RE, webpage))
 491         ]
 492
 493         playlist_title = self._html_search_regex(
 494             self._PLAYLIST_TITLE_RE, webpage, 'playlist title',
 495             default=None) or self._og_search_title(webpage)
 496
 497         return self.playlist_result(entries, playlist_id, playlist_title)
 498
 499
 500 class VPROIE(NPOPlaylistBaseIE):
 501     IE_NAME = 'vpro'
 502     _VALID_URL = r'https?://(?:www\.)?(?:(?:tegenlicht\.)?vpro|2doc)\.nl/(?:[^/]+/)*(?P<id>[^/]+)\.html'
 503     _PLAYLIST_TITLE_RE = (r'<h1[^>]+class=["\'].*?\bmedia-platform-title\b.*?["\'][^>]*>([^<]+)',
 504                           r'<h5[^>]+class=["\'].*?\bmedia-platform-subtitle\b.*?["\'][^>]*>([^<]+)')
 505     _PLAYLIST_ENTRY_RE = r'data-media-id="([^"]+)"'
 506
 507     _TESTS = [
 508         {
 509             'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
 510             'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
 511             'info_dict': {
 512                 'id': 'VPWON_1169289',
 513                 'ext': 'm4v',
 514                 'title': 'De toekomst komt uit Afrika',
 515                 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
 516                 'upload_date': '20130225',
 517             },
 518             'skip': 'Video gone',
 519         },
 520         {
 521             'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html',
 522             'info_dict': {
 523                 'id': 'sergio-herman',
 524                 'title': 'sergio herman: fucking perfect',
 525             },
 526             'playlist_count': 2,
 527         },
 528         {
 529             # playlist with youtube embed
 530             'url': 'http://www.vpro.nl/programmas/2doc/2015/education-education.html',
 531             'info_dict': {
 532                 'id': 'education-education',
 533                 'title': 'education education',
 534             },
 535             'playlist_count': 2,
 536         },
 537         {
 538             'url': 'http://www.2doc.nl/documentaires/series/2doc/2015/oktober/de-tegenprestatie.html',
 539             'info_dict': {
 540                 'id': 'de-tegenprestatie',
 541                 'title': 'De Tegenprestatie',
 542             },
 543             'playlist_count': 2,
 544         }, {
 545             'url': 'http://www.2doc.nl/speel~VARA_101375237~mh17-het-verdriet-van-nederland~.html',
 546             'info_dict': {
 547                 'id': 'VARA_101375237',
 548                 'ext': 'm4v',
 549                 'title': 'MH17: Het verdriet van Nederland',
 550                 'description': 'md5:09e1a37c1fdb144621e22479691a9f18',
 551                 'upload_date': '20150716',
 552             },
 553             'params': {
 554                 # Skip because of m3u8 download
 555                 'skip_download': True
 556             },
 557         }
 558     ]
 559
 560
 561 class WNLIE(NPOPlaylistBaseIE):
 562     IE_NAME = 'wnl'
 563     _VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+'
 564     _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>'
 565     _PLAYLIST_ENTRY_RE = r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>Deel \d+'
 566
 567     _TESTS = [{
 568         'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515',
 569         'info_dict': {
 570             'id': 'vandaag-de-dag-6-mei',
 571             'title': 'Vandaag de Dag 6 mei',
 572         },
 573         'playlist_count': 4,
 574     }]
 575
 576
 577 class AndereTijdenIE(NPOPlaylistBaseIE):
 578     IE_NAME = 'anderetijden'
 579     _VALID_URL = r'https?://(?:www\.)?anderetijden\.nl/programma/(?:[^/]+/)+(?P<id>[^/?#&]+)'
 580     _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class=["\'].*?\bpage-title\b.*?["\'][^>]*>(.+?)</h1>'
 581     _PLAYLIST_ENTRY_RE = r'<figure[^>]+class=["\']episode-container episode-page["\'][^>]+data-prid=["\'](.+?)["\']'
 582
 583     _TESTS = [{
 584         'url': 'http://anderetijden.nl/programma/1/Andere-Tijden/aflevering/676/Duitse-soldaten-over-de-Slag-bij-Arnhem',
 585         'info_dict': {
 586             'id': 'Duitse-soldaten-over-de-Slag-bij-Arnhem',
 587             'title': 'Duitse soldaten over de Slag bij Arnhem',
 588         },
 589         'playlist_count': 3,
 590     }]