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