_ Git - youtube-dl/blob - youtube_dl/extractor/nrk.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     ExtractorError,
  10     float_or_none,
  11     parse_duration,
  12     unified_strdate,
  13 )
  14
  15
  16 class NRKIE(InfoExtractor):
  17     _VALID_URL = r'(?:nrk:|http://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
  18
  19     _TESTS = [
  20         {
  21             'url': 'http://www.nrk.no/video/PS*150533',
  22             'md5': 'bccd850baebefe23b56d708a113229c2',
  23             'info_dict': {
  24                 'id': '150533',
  25                 'ext': 'flv',
  26                 'title': 'Dompap og andre fugler i Piip-Show',
  27                 'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
  28                 'duration': 263,
  29             }
  30         },
  31         {
  32             'url': 'http://www.nrk.no/video/PS*154915',
  33             'md5': '0b1493ba1aae7d9579a5ad5531bc395a',
  34             'info_dict': {
  35                 'id': '154915',
  36                 'ext': 'flv',
  37                 'title': 'Slik høres internett ut når du er blind',
  38                 'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
  39                 'duration': 20,
  40             }
  41         },
  42     ]
  43
  44     def _real_extract(self, url):
  45         video_id = self._match_id(url)
  46
  47         data = self._download_json(
  48             'http://v8.psapi.nrk.no/mediaelement/%s' % video_id,
  49             video_id, 'Downloading media JSON')
  50
  51         if data['usageRights']['isGeoBlocked']:
  52             raise ExtractorError(
  53                 'NRK har ikke rettig-heter til å vise dette programmet utenfor Norge',
  54                 expected=True)
  55
  56         video_url = data['mediaUrl'] + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81'
  57
  58         duration = parse_duration(data.get('duration'))
  59
  60         images = data.get('images')
  61         if images:
  62             thumbnails = images['webImages']
  63             thumbnails.sort(key=lambda image: image['pixelWidth'])
  64             thumbnail = thumbnails[-1]['imageUrl']
  65         else:
  66             thumbnail = None
  67
  68         return {
  69             'id': video_id,
  70             'url': video_url,
  71             'ext': 'flv',
  72             'title': data['title'],
  73             'description': data['description'],
  74             'duration': duration,
  75             'thumbnail': thumbnail,
  76         }
  77
  78
  79 class NRKPlaylistIE(InfoExtractor):
  80     _VALID_URL = r'http://(?:www\.)?nrk\.no/(?:[^/]+/)*(?P<id>[^/]+)'
  81
  82     _TEST = {
  83         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
  84         'info_dict': {
  85             'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
  86             'title': 'Gjenopplev den historiske solformørkelsen',
  87             'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
  88         },
  89         'playlist_mincount': 2,
  90     }
  91
  92     def _real_extract(self, url):
  93         playlist_id = self._match_id(url)
  94
  95         webpage = self._download_webpage(url, playlist_id)
  96
  97         entries = [
  98             self.url_result('nrk:%s' % video_id, 'NRK')
  99             for video_id in re.findall(
 100                 r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="(\d+)"', webpage)
 101         ]
 102
 103         playlist_title = self._og_search_title(webpage)
 104         playlist_description = self._og_search_description(webpage)
 105
 106         return self.playlist_result(
 107             entries, playlist_id, playlist_title, playlist_description)
 108
 109
 110 class NRKTVIE(InfoExtractor):
 111     _VALID_URL = r'(?P<baseurl>http://tv\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
 112
 113     _TESTS = [
 114         {
 115             'url': 'http://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 116             'md5': 'adf2c5454fa2bf032f47a9f8fb351342',
 117             'info_dict': {
 118                 'id': 'MUHH48000314',
 119                 'ext': 'flv',
 120                 'title': '20 spørsmål',
 121                 'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 122                 'upload_date': '20140523',
 123                 'duration': 1741.52,
 124             },
 125         },
 126         {
 127             'url': 'http://tv.nrk.no/program/mdfp15000514',
 128             'md5': '383650ece2b25ecec996ad7b5bb2a384',
 129             'info_dict': {
 130                 'id': 'mdfp15000514',
 131                 'ext': 'flv',
 132                 'title': 'Kunnskapskanalen: Grunnlovsjubiléet - Stor ståhei for ingenting',
 133                 'description': 'md5:654c12511f035aed1e42bdf5db3b206a',
 134                 'upload_date': '20140524',
 135                 'duration': 4605.0,
 136             },
 137         },
 138         {
 139             # single playlist video
 140             'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 141             'md5': 'adbd1dbd813edaf532b0a253780719c2',
 142             'info_dict': {
 143                 'id': 'MSPO40010515-part2',
 144                 'ext': 'flv',
 145                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 146                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 147                 'upload_date': '20150106',
 148             },
 149             'skip': 'Only works from Norway',
 150         },
 151         {
 152             'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 153             'playlist': [
 154                 {
 155                     'md5': '9480285eff92d64f06e02a5367970a7a',
 156                     'info_dict': {
 157                         'id': 'MSPO40010515-part1',
 158                         'ext': 'flv',
 159                         'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
 160                         'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 161                         'upload_date': '20150106',
 162                     },
 163                 },
 164                 {
 165                     'md5': 'adbd1dbd813edaf532b0a253780719c2',
 166                     'info_dict': {
 167                         'id': 'MSPO40010515-part2',
 168                         'ext': 'flv',
 169                         'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 170                         'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 171                         'upload_date': '20150106',
 172                     },
 173                 },
 174             ],
 175             'info_dict': {
 176                 'id': 'MSPO40010515',
 177                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
 178                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 179                 'upload_date': '20150106',
 180                 'duration': 6947.5199999999995,
 181             },
 182             'skip': 'Only works from Norway',
 183         }
 184     ]
 185
 186     def _debug_print(self, txt):
 187         if self._downloader.params.get('verbose', False):
 188             self.to_screen('[debug] %s' % txt)
 189
 190     def _get_subtitles(self, subtitlesurl, video_id, baseurl):
 191         url = "%s%s" % (baseurl, subtitlesurl)
 192         self._debug_print('%s: Subtitle url: %s' % (video_id, url))
 193         captions = self._download_xml(
 194             url, video_id, 'Downloading subtitles',
 195             transform_source=lambda s: s.replace(r'<br />', '\r\n'))
 196         lang = captions.get('lang', 'no')
 197         ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/ns/ttml}'))
 198         srt = ''
 199         for pos, p in enumerate(ps):
 200             begin = parse_duration(p.get('begin'))
 201             duration = parse_duration(p.get('dur'))
 202             starttime = self._subtitles_timecode(begin)
 203             endtime = self._subtitles_timecode(begin + duration)
 204             srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (compat_str(pos), starttime, endtime, p.text)
 205         return {lang: [
 206             {'ext': 'ttml', 'url': url},
 207             {'ext': 'srt', 'data': srt},
 208         ]}
 209
 210     def _extract_f4m(self, manifest_url, video_id):
 211         return self._extract_f4m_formats(manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id)
 212
 213     def _real_extract(self, url):
 214         mobj = re.match(self._VALID_URL, url)
 215         video_id = mobj.group('id')
 216         part_id = mobj.group('part_id')
 217         baseurl = mobj.group('baseurl')
 218
 219         webpage = self._download_webpage(url, video_id)
 220
 221         title = self._html_search_meta(
 222             'title', webpage, 'title')
 223         description = self._html_search_meta(
 224             'description', webpage, 'description')
 225
 226         thumbnail = self._html_search_regex(
 227             r'data-posterimage="([^"]+)"',
 228             webpage, 'thumbnail', fatal=False)
 229         upload_date = unified_strdate(self._html_search_meta(
 230             'rightsfrom', webpage, 'upload date', fatal=False))
 231         duration = float_or_none(self._html_search_regex(
 232             r'data-duration="([^"]+)"',
 233             webpage, 'duration', fatal=False))
 234
 235         # playlist
 236         parts = re.findall(
 237             r'<a href="#del=(\d+)"[^>]+data-argument="([^"]+)">([^<]+)</a>', webpage)
 238         if parts:
 239             entries = []
 240             for current_part_id, stream_url, part_title in parts:
 241                 if part_id and current_part_id != part_id:
 242                     continue
 243                 video_part_id = '%s-part%s' % (video_id, current_part_id)
 244                 formats = self._extract_f4m(stream_url, video_part_id)
 245                 entries.append({
 246                     'id': video_part_id,
 247                     'title': part_title,
 248                     'description': description,
 249                     'thumbnail': thumbnail,
 250                     'upload_date': upload_date,
 251                     'formats': formats,
 252                 })
 253             if part_id:
 254                 if entries:
 255                     return entries[0]
 256             else:
 257                 playlist = self.playlist_result(entries, video_id, title, description)
 258                 playlist.update({
 259                     'thumbnail': thumbnail,
 260                     'upload_date': upload_date,
 261                     'duration': duration,
 262                 })
 263                 return playlist
 264
 265         formats = []
 266
 267         f4m_url = re.search(r'data-media="([^"]+)"', webpage)
 268         if f4m_url:
 269             formats.extend(self._extract_f4m(f4m_url.group(1), video_id))
 270
 271         m3u8_url = re.search(r'data-hls-media="([^"]+)"', webpage)
 272         if m3u8_url:
 273             formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4'))
 274         self._sort_formats(formats)
 275
 276         subtitles_url = self._html_search_regex(
 277             r'data-subtitlesurl[ ]*=[ ]*"([^"]+)"',
 278             webpage, 'subtitle URL', default=None)
 279         subtitles = None
 280         if subtitles_url:
 281             subtitles = self.extract_subtitles(subtitles_url, video_id, baseurl)
 282
 283         return {
 284             'id': video_id,
 285             'title': title,
 286             'description': description,
 287             'thumbnail': thumbnail,
 288             'upload_date': upload_date,
 289             'duration': duration,
 290             'formats': formats,
 291             'subtitles': subtitles,
 292         }