Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/npo.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     fix_xml_ampersands,
   6     parse_duration,
   7     qualities,
   8     strip_jsonp,
   9     unified_strdate,
  10     url_basename,
  11 )
  12
  13
  14 class NPOBaseIE(InfoExtractor):
  15     def _get_token(self, video_id):
  16         token_page = self._download_webpage(
  17             'http://ida.omroep.nl/npoplayer/i.js',
  18             video_id, note='Downloading token')
  19         return self._search_regex(
  20             r'npoplayer\.token = "(.+?)"', token_page, 'token')
  21
  22
  23 class NPOIE(NPOBaseIE):
  24     IE_NAME = 'npo.nl'
  25     _VALID_URL = r'https?://(?:www\.)?npo\.nl/(?!live|radio)[^/]+/[^/]+/(?P<id>[^/?]+)'
  26
  27     _TESTS = [
  28         {
  29             'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
  30             'md5': '4b3f9c429157ec4775f2c9cb7b911016',
  31             'info_dict': {
  32                 'id': 'VPWON_1220719',
  33                 'ext': 'm4v',
  34                 'title': 'Nieuwsuur',
  35                 'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
  36                 'upload_date': '20140622',
  37             },
  38         },
  39         {
  40             'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
  41             'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
  42             'info_dict': {
  43                 'id': 'VARA_101191800',
  44                 'ext': 'm4v',
  45                 'title': 'De Mega Mike & Mega Thomas show',
  46                 'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
  47                 'upload_date': '20090227',
  48                 'duration': 2400,
  49             },
  50         },
  51         {
  52             'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
  53             'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
  54             'info_dict': {
  55                 'id': 'VPWON_1169289',
  56                 'ext': 'm4v',
  57                 'title': 'Tegenlicht',
  58                 'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
  59                 'upload_date': '20130225',
  60                 'duration': 3000,
  61             },
  62         },
  63         {
  64             'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
  65             'info_dict': {
  66                 'id': 'WO_VPRO_043706',
  67                 'ext': 'wmv',
  68                 'title': 'De nieuwe mens - Deel 1',
  69                 'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
  70                 'duration': 4680,
  71             },
  72             'params': {
  73                 # mplayer mms download
  74                 'skip_download': True,
  75             }
  76         },
  77         # non asf in streams
  78         {
  79             'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
  80             'md5': 'b3da13de374cbe2d5332a7e910bef97f',
  81             'info_dict': {
  82                 'id': 'WO_NOS_762771',
  83                 'ext': 'mp4',
  84                 'title': 'Hoe gaat Europa verder na Parijs?',
  85             },
  86         },
  87     ]
  88
  89     def _real_extract(self, url):
  90         video_id = self._match_id(url)
  91         return self._get_info(video_id)
  92
  93     def _get_info(self, video_id):
  94         metadata = self._download_json(
  95             'http://e.omroep.nl/metadata/aflevering/%s' % video_id,
  96             video_id,
  97             # We have to remove the javascript callback
  98             transform_source=strip_jsonp,
  99         )
 100
 101         token = self._get_token(video_id)
 102
 103         formats = []
 104
 105         pubopties = metadata.get('pubopties')
 106         if pubopties:
 107             quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std'])
 108             for format_id in pubopties:
 109                 format_info = self._download_json(
 110                     'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s'
 111                     % (video_id, format_id, token),
 112                     video_id, 'Downloading %s JSON' % format_id)
 113                 if format_info.get('error_code', 0) or format_info.get('errorcode', 0):
 114                     continue
 115                 streams = format_info.get('streams')
 116                 if streams:
 117                     video_info = self._download_json(
 118                         streams[0] + '&type=json',
 119                         video_id, 'Downloading %s stream JSON' % format_id)
 120                 else:
 121                     video_info = format_info
 122                 video_url = video_info.get('url')
 123                 if not video_url:
 124                     continue
 125                 if format_id == 'adaptive':
 126                     formats.extend(self._extract_m3u8_formats(video_url, video_id))
 127                 else:
 128                     formats.append({
 129                         'url': video_url,
 130                         'format_id': format_id,
 131                         'quality': quality(format_id),
 132                     })
 133
 134         streams = metadata.get('streams')
 135         if streams:
 136             for i, stream in enumerate(streams):
 137                 stream_url = stream.get('url')
 138                 if not stream_url:
 139                     continue
 140                 if '.asf' not in stream_url:
 141                     formats.append({
 142                         'url': stream_url,
 143                         'quality': stream.get('kwaliteit'),
 144                     })
 145                     continue
 146                 asx = self._download_xml(
 147                     stream_url, video_id,
 148                     'Downloading stream %d ASX playlist' % i,
 149                     transform_source=fix_xml_ampersands)
 150                 ref = asx.find('./ENTRY/Ref')
 151                 if ref is None:
 152                     continue
 153                 video_url = ref.get('href')
 154                 if not video_url:
 155                     continue
 156                 formats.append({
 157                     'url': video_url,
 158                     'ext': stream.get('formaat', 'asf'),
 159                     'quality': stream.get('kwaliteit'),
 160                 })
 161
 162         self._sort_formats(formats)
 163
 164         subtitles = {}
 165         if metadata.get('tt888') == 'ja':
 166             subtitles['nl'] = [{
 167                 'ext': 'vtt',
 168                 'url': 'http://e.omroep.nl/tt888/%s' % video_id,
 169             }]
 170
 171         return {
 172             'id': video_id,
 173             'title': metadata['titel'],
 174             'description': metadata['info'],
 175             'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
 176             'upload_date': unified_strdate(metadata.get('gidsdatum')),
 177             'duration': parse_duration(metadata.get('tijdsduur')),
 178             'formats': formats,
 179             'subtitles': subtitles,
 180         }
 181
 182
 183 class NPOLiveIE(NPOBaseIE):
 184     IE_NAME = 'npo.nl:live'
 185     _VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P<id>.+)'
 186
 187     _TEST = {
 188         'url': 'http://www.npo.nl/live/npo-1',
 189         'info_dict': {
 190             'id': 'LI_NEDERLAND1_136692',
 191             'display_id': 'npo-1',
 192             'ext': 'mp4',
 193             'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 194             'description': 'Livestream',
 195             'is_live': True,
 196         },
 197         'params': {
 198             'skip_download': True,
 199         }
 200     }
 201
 202     def _real_extract(self, url):
 203         display_id = self._match_id(url)
 204
 205         webpage = self._download_webpage(url, display_id)
 206
 207         live_id = self._search_regex(
 208             r'data-prid="([^"]+)"', webpage, 'live id')
 209
 210         metadata = self._download_json(
 211             'http://e.omroep.nl/metadata/%s' % live_id,
 212             display_id, transform_source=strip_jsonp)
 213
 214         token = self._get_token(display_id)
 215
 216         formats = []
 217
 218         streams = metadata.get('streams')
 219         if streams:
 220             for stream in streams:
 221                 stream_type = stream.get('type').lower()
 222                 # smooth streaming is not supported
 223                 if stream_type in ['ss', 'ms']:
 224                     continue
 225                 stream_info = self._download_json(
 226                     'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp'
 227                     % (stream.get('url'), token),
 228                     display_id, 'Downloading %s JSON' % stream_type)
 229                 if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0):
 230                     continue
 231                 stream_url = self._download_json(
 232                     stream_info['stream'], display_id,
 233                     'Downloading %s URL' % stream_type,
 234                     'Unable to download %s URL' % stream_type,
 235                     transform_source=strip_jsonp, fatal=False)
 236                 if not stream_url:
 237                     continue
 238                 if stream_type == 'hds':
 239                     f4m_formats = self._extract_f4m_formats(stream_url, display_id)
 240                     # f4m downloader downloads only piece of live stream
 241                     for f4m_format in f4m_formats:
 242                         f4m_format['preference'] = -1
 243                     formats.extend(f4m_formats)
 244                 elif stream_type == 'hls':
 245                     formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4'))
 246                 else:
 247                     formats.append({
 248                         'url': stream_url,
 249                         'preference': -10,
 250                     })
 251
 252         self._sort_formats(formats)
 253
 254         return {
 255             'id': live_id,
 256             'display_id': display_id,
 257             'title': self._live_title(metadata['titel']),
 258             'description': metadata['info'],
 259             'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
 260             'formats': formats,
 261             'is_live': True,
 262         }
 263
 264
 265 class NPORadioIE(InfoExtractor):
 266     IE_NAME = 'npo.nl:radio'
 267     _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
 268
 269     _TEST = {
 270         'url': 'http://www.npo.nl/radio/radio-1',
 271         'info_dict': {
 272             'id': 'radio-1',
 273             'ext': 'mp3',
 274             'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 275             'is_live': True,
 276         },
 277         'params': {
 278             'skip_download': True,
 279         }
 280     }
 281
 282     @staticmethod
 283     def _html_get_attribute_regex(attribute):
 284         return r'{0}\s*=\s*\'([^\']+)\''.format(attribute)
 285
 286     def _real_extract(self, url):
 287         video_id = self._match_id(url)
 288
 289         webpage = self._download_webpage(url, video_id)
 290
 291         title = self._html_search_regex(
 292             self._html_get_attribute_regex('data-channel'), webpage, 'title')
 293
 294         stream = self._parse_json(
 295             self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
 296             video_id)
 297
 298         codec = stream.get('codec')
 299
 300         return {
 301             'id': video_id,
 302             'url': stream['url'],
 303             'title': self._live_title(title),
 304             'acodec': codec,
 305             'ext': codec,
 306             'is_live': True,
 307         }
 308
 309
 310 class NPORadioFragmentIE(InfoExtractor):
 311     IE_NAME = 'npo.nl:radio:fragment'
 312     _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
 313
 314     _TEST = {
 315         'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
 316         'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
 317         'info_dict': {
 318             'id': '174356',
 319             'ext': 'mp3',
 320             'title': 'Jubileumconcert Willeke Alberti',
 321         },
 322     }
 323
 324     def _real_extract(self, url):
 325         audio_id = self._match_id(url)
 326
 327         webpage = self._download_webpage(url, audio_id)
 328
 329         title = self._html_search_regex(
 330             r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
 331             webpage, 'title')
 332
 333         audio_url = self._search_regex(
 334             r"data-streams='([^']+)'", webpage, 'audio url')
 335
 336         return {
 337             'id': audio_id,
 338             'url': audio_url,
 339             'title': title,
 340         }
 341
 342
 343 class TegenlichtVproIE(NPOIE):
 344     IE_NAME = 'tegenlicht.vpro.nl'
 345     _VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?'
 346
 347     _TESTS = [
 348         {
 349             'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
 350             'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
 351             'info_dict': {
 352                 'id': 'VPWON_1169289',
 353                 'ext': 'm4v',
 354                 'title': 'Tegenlicht',
 355                 'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
 356                 'upload_date': '20130225',
 357             },
 358         },
 359     ]
 360
 361     def _real_extract(self, url):
 362         name = url_basename(url)
 363         webpage = self._download_webpage(url, name)
 364         urn = self._html_search_meta('mediaurn', webpage)
 365         info_page = self._download_json(
 366             'http://rs.vpro.nl/v2/api/media/%s.json' % urn, name)
 367         return self._get_info(info_page['mid'])