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