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 (
   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 and ntr.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\.nl/[^/]+/[^/]+/
  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         # live stream
 151         'url': 'npo:LI_NL1_4188102',
 152         'only_matching': True,
 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         urls = set()
 183
 184         quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std'])
 185         items = self._download_json(
 186             'http://ida.omroep.nl/app.php/%s' % video_id, video_id,
 187             'Downloading formats JSON', query={
 188                 'adaptive': 'yes',
 189                 'token': token,
 190             })['items'][0]
 191         for num, item in enumerate(items):
 192             item_url = item.get('url')
 193             if not item_url or item_url in urls:
 194                 continue
 195             urls.add(item_url)
 196             format_id = self._search_regex(
 197                 r'video/ida/([^/]+)', item_url, 'format id',
 198                 default=None)
 199
 200             def add_format_url(format_url):
 201                 formats.append({
 202                     'url': format_url,
 203                     'format_id': format_id,
 204                     'quality': quality(format_id),
 205                 })
 206
 207             # Example: http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706
 208             if item.get('contentType') in ('url', 'audio'):
 209                 add_format_url(item_url)
 210                 continue
 211
 212             try:
 213                 stream_info = self._download_json(
 214                     item_url + '&type=json', video_id,
 215                     'Downloading %s stream JSON'
 216                     % item.get('label') or item.get('format') or format_id or num)
 217             except ExtractorError as ee:
 218                 if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
 219                     error = (self._parse_json(
 220                         ee.cause.read().decode(), video_id,
 221                         fatal=False) or {}).get('errorstring')
 222                     if error:
 223                         raise ExtractorError(error, expected=True)
 224                 raise
 225             # Stream URL instead of JSON, example: npo:LI_NL1_4188102
 226             if isinstance(stream_info, compat_str):
 227                 if not stream_info.startswith('http'):
 228                     continue
 229                 video_url = stream_info
 230             # JSON
 231             else:
 232                 video_url = stream_info.get('url')
 233             if not video_url or video_url in urls:
 234                 continue
 235             urls.add(item_url)
 236             if determine_ext(video_url) == 'm3u8':
 237                 formats.extend(self._extract_m3u8_formats(
 238                     video_url, video_id, ext='mp4',
 239                     entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
 240             else:
 241                 add_format_url(video_url)
 242
 243         is_live = metadata.get('medium') == 'live'
 244
 245         if not is_live:
 246             for num, stream in enumerate(metadata.get('streams', [])):
 247                 stream_url = stream.get('url')
 248                 if not stream_url or stream_url in urls:
 249                     continue
 250                 urls.add(stream_url)
 251                 # smooth streaming is not supported
 252                 stream_type = stream.get('type', '').lower()
 253                 if stream_type in ['ss', 'ms']:
 254                     continue
 255                 if stream_type == 'hds':
 256                     f4m_formats = self._extract_f4m_formats(
 257                         stream_url, video_id, fatal=False)
 258                     # f4m downloader downloads only piece of live stream
 259                     for f4m_format in f4m_formats:
 260                         f4m_format['preference'] = -1
 261                     formats.extend(f4m_formats)
 262                 elif stream_type == 'hls':
 263                     formats.extend(self._extract_m3u8_formats(
 264                         stream_url, video_id, ext='mp4', fatal=False))
 265                 # Example: http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706
 266                 elif '.asf' in stream_url:
 267                     asx = self._download_xml(
 268                         stream_url, video_id,
 269                         'Downloading stream %d ASX playlist' % num,
 270                         transform_source=fix_xml_ampersands, fatal=False)
 271                     if not asx:
 272                         continue
 273                     ref = asx.find('./ENTRY/Ref')
 274                     if ref is None:
 275                         continue
 276                     video_url = ref.get('href')
 277                     if not video_url or video_url in urls:
 278                         continue
 279                     urls.add(video_url)
 280                     formats.append({
 281                         'url': video_url,
 282                         'ext': stream.get('formaat', 'asf'),
 283                         'quality': stream.get('kwaliteit'),
 284                         'preference': -10,
 285                     })
 286                 else:
 287                     formats.append({
 288                         'url': stream_url,
 289                         'quality': stream.get('kwaliteit'),
 290                     })
 291
 292         self._sort_formats(formats)
 293
 294         subtitles = {}
 295         if metadata.get('tt888') == 'ja':
 296             subtitles['nl'] = [{
 297                 'ext': 'vtt',
 298                 'url': 'http://tt888.omroep.nl/tt888/%s' % video_id,
 299             }]
 300
 301         return {
 302             'id': video_id,
 303             'title': self._live_title(title) if is_live else title,
 304             'description': metadata.get('info'),
 305             'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
 306             'upload_date': unified_strdate(metadata.get('gidsdatum')),
 307             'duration': parse_duration(metadata.get('tijdsduur')),
 308             'formats': formats,
 309             'subtitles': subtitles,
 310             'is_live': is_live,
 311         }
 312
 313
 314 class NPOLiveIE(NPOBaseIE):
 315     IE_NAME = 'npo.nl:live'
 316     _VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P<id>[^/?#&]+)'
 317
 318     _TEST = {
 319         'url': 'http://www.npo.nl/live/npo-1',
 320         'info_dict': {
 321             'id': 'LI_NL1_4188102',
 322             'display_id': 'npo-1',
 323             'ext': 'mp4',
 324             'title': 're:^NPO 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 325             'is_live': True,
 326         },
 327         'params': {
 328             'skip_download': True,
 329         }
 330     }
 331
 332     def _real_extract(self, url):
 333         display_id = self._match_id(url)
 334
 335         webpage = self._download_webpage(url, display_id)
 336
 337         live_id = self._search_regex(
 338             r'data-prid="([^"]+)"', webpage, 'live id')
 339
 340         return {
 341             '_type': 'url_transparent',
 342             'url': 'npo:%s' % live_id,
 343             'ie_key': NPOIE.ie_key(),
 344             'id': live_id,
 345             'display_id': display_id,
 346         }
 347
 348
 349 class NPORadioIE(InfoExtractor):
 350     IE_NAME = 'npo.nl:radio'
 351     _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
 352
 353     _TEST = {
 354         'url': 'http://www.npo.nl/radio/radio-1',
 355         'info_dict': {
 356             'id': 'radio-1',
 357             'ext': 'mp3',
 358             'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 359             'is_live': True,
 360         },
 361         'params': {
 362             'skip_download': True,
 363         }
 364     }
 365
 366     @staticmethod
 367     def _html_get_attribute_regex(attribute):
 368         return r'{0}\s*=\s*\'([^\']+)\''.format(attribute)
 369
 370     def _real_extract(self, url):
 371         video_id = self._match_id(url)
 372
 373         webpage = self._download_webpage(url, video_id)
 374
 375         title = self._html_search_regex(
 376             self._html_get_attribute_regex('data-channel'), webpage, 'title')
 377
 378         stream = self._parse_json(
 379             self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
 380             video_id)
 381
 382         codec = stream.get('codec')
 383
 384         return {
 385             'id': video_id,
 386             'url': stream['url'],
 387             'title': self._live_title(title),
 388             'acodec': codec,
 389             'ext': codec,
 390             'is_live': True,
 391         }
 392
 393
 394 class NPORadioFragmentIE(InfoExtractor):
 395     IE_NAME = 'npo.nl:radio:fragment'
 396     _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
 397
 398     _TEST = {
 399         'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
 400         'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
 401         'info_dict': {
 402             'id': '174356',
 403             'ext': 'mp3',
 404             'title': 'Jubileumconcert Willeke Alberti',
 405         },
 406     }
 407
 408     def _real_extract(self, url):
 409         audio_id = self._match_id(url)
 410
 411         webpage = self._download_webpage(url, audio_id)
 412
 413         title = self._html_search_regex(
 414             r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
 415             webpage, 'title')
 416
 417         audio_url = self._search_regex(
 418             r"data-streams='([^']+)'", webpage, 'audio url')
 419
 420         return {
 421             'id': audio_id,
 422             'url': audio_url,
 423             'title': title,
 424         }
 425
 426
 427 class NPODataMidEmbedIE(InfoExtractor):
 428     def _real_extract(self, url):
 429         display_id = self._match_id(url)
 430         webpage = self._download_webpage(url, display_id)
 431         video_id = self._search_regex(
 432             r'data-mid=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video_id', group='id')
 433         return {
 434             '_type': 'url_transparent',
 435             'ie_key': 'NPO',
 436             'url': 'npo:%s' % video_id,
 437             'display_id': display_id
 438         }
 439
 440
 441 class SchoolTVIE(NPODataMidEmbedIE):
 442     IE_NAME = 'schooltv'
 443     _VALID_URL = r'https?://(?:www\.)?schooltv\.nl/video/(?P<id>[^/?#&]+)'
 444
 445     _TEST = {
 446         'url': 'http://www.schooltv.nl/video/ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam/',
 447         'info_dict': {
 448             'id': 'WO_NTR_429477',
 449             'display_id': 'ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam',
 450             'title': 'Ademhaling: De hele dag haal je adem. Maar wat gebeurt er dan eigenlijk in je lichaam?',
 451             'ext': 'mp4',
 452             'description': 'md5:abfa0ff690adb73fd0297fd033aaa631'
 453         },
 454         'params': {
 455             # Skip because of m3u8 download
 456             'skip_download': True
 457         }
 458     }
 459
 460
 461 class HetKlokhuisIE(NPODataMidEmbedIE):
 462     IE_NAME = 'hetklokhuis'
 463     _VALID_URL = r'https?://(?:www\.)?hetklokhuis.nl/[^/]+/\d+/(?P<id>[^/?#&]+)'
 464
 465     _TEST = {
 466         'url': 'http://hetklokhuis.nl/tv-uitzending/3471/Zwaartekrachtsgolven',
 467         'info_dict': {
 468             'id': 'VPWON_1260528',
 469             'display_id': 'Zwaartekrachtsgolven',
 470             'ext': 'm4v',
 471             'title': 'Het Klokhuis: Zwaartekrachtsgolven',
 472             'description': 'md5:c94f31fb930d76c2efa4a4a71651dd48',
 473             'upload_date': '20170223',
 474         },
 475         'params': {
 476             'skip_download': True
 477         }
 478     }
 479
 480
 481 class NPOPlaylistBaseIE(NPOIE):
 482     def _real_extract(self, url):
 483         playlist_id = self._match_id(url)
 484
 485         webpage = self._download_webpage(url, playlist_id)
 486
 487         entries = [
 488             self.url_result('npo:%s' % video_id if not video_id.startswith('http') else video_id)
 489             for video_id in orderedSet(re.findall(self._PLAYLIST_ENTRY_RE, webpage))
 490         ]
 491
 492         playlist_title = self._html_search_regex(
 493             self._PLAYLIST_TITLE_RE, webpage, 'playlist title',
 494             default=None) or self._og_search_title(webpage)
 495
 496         return self.playlist_result(entries, playlist_id, playlist_title)
 497
 498
 499 class VPROIE(NPOPlaylistBaseIE):
 500     IE_NAME = 'vpro'
 501     _VALID_URL = r'https?://(?:www\.)?(?:(?:tegenlicht\.)?vpro|2doc)\.nl/(?:[^/]+/)*(?P<id>[^/]+)\.html'
 502     _PLAYLIST_TITLE_RE = (r'<h1[^>]+class=["\'].*?\bmedia-platform-title\b.*?["\'][^>]*>([^<]+)',
 503                           r'<h5[^>]+class=["\'].*?\bmedia-platform-subtitle\b.*?["\'][^>]*>([^<]+)')
 504     _PLAYLIST_ENTRY_RE = r'data-media-id="([^"]+)"'
 505
 506     _TESTS = [
 507         {
 508             'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
 509             'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
 510             'info_dict': {
 511                 'id': 'VPWON_1169289',
 512                 'ext': 'm4v',
 513                 'title': 'De toekomst komt uit Afrika',
 514                 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
 515                 'upload_date': '20130225',
 516             },
 517             'skip': 'Video gone',
 518         },
 519         {
 520             'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html',
 521             'info_dict': {
 522                 'id': 'sergio-herman',
 523                 'title': 'sergio herman: fucking perfect',
 524             },
 525             'playlist_count': 2,
 526         },
 527         {
 528             # playlist with youtube embed
 529             'url': 'http://www.vpro.nl/programmas/2doc/2015/education-education.html',
 530             'info_dict': {
 531                 'id': 'education-education',
 532                 'title': 'education education',
 533             },
 534             'playlist_count': 2,
 535         },
 536         {
 537             'url': 'http://www.2doc.nl/documentaires/series/2doc/2015/oktober/de-tegenprestatie.html',
 538             'info_dict': {
 539                 'id': 'de-tegenprestatie',
 540                 'title': 'De Tegenprestatie',
 541             },
 542             'playlist_count': 2,
 543         }, {
 544             'url': 'http://www.2doc.nl/speel~VARA_101375237~mh17-het-verdriet-van-nederland~.html',
 545             'info_dict': {
 546                 'id': 'VARA_101375237',
 547                 'ext': 'm4v',
 548                 'title': 'MH17: Het verdriet van Nederland',
 549                 'description': 'md5:09e1a37c1fdb144621e22479691a9f18',
 550                 'upload_date': '20150716',
 551             },
 552             'params': {
 553                 # Skip because of m3u8 download
 554                 'skip_download': True
 555             },
 556         }
 557     ]
 558
 559
 560 class WNLIE(NPOPlaylistBaseIE):
 561     IE_NAME = 'wnl'
 562     _VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+'
 563     _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>'
 564     _PLAYLIST_ENTRY_RE = r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>Deel \d+'
 565
 566     _TESTS = [{
 567         'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515',
 568         'info_dict': {
 569             'id': 'vandaag-de-dag-6-mei',
 570             'title': 'Vandaag de Dag 6 mei',
 571         },
 572         'playlist_count': 4,
 573     }]
 574
 575
 576 class AndereTijdenIE(NPOPlaylistBaseIE):
 577     IE_NAME = 'anderetijden'
 578     _VALID_URL = r'https?://(?:www\.)?anderetijden\.nl/programma/(?:[^/]+/)+(?P<id>[^/?#&]+)'
 579     _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class=["\'].*?\bpage-title\b.*?["\'][^>]*>(.+?)</h1>'
 580     _PLAYLIST_ENTRY_RE = r'<figure[^>]+class=["\']episode-container episode-page["\'][^>]+data-prid=["\'](.+?)["\']'
 581
 582     _TESTS = [{
 583         'url': 'http://anderetijden.nl/programma/1/Andere-Tijden/aflevering/676/Duitse-soldaten-over-de-Slag-bij-Arnhem',
 584         'info_dict': {
 585             'id': 'Duitse-soldaten-over-de-Slag-bij-Arnhem',
 586             'title': 'Duitse soldaten over de Slag bij Arnhem',
 587         },
 588         'playlist_count': 3,
 589     }]