_ 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/(?!video)(?:[^/]+/)+(?P<id>[^/]+)'
  81
  82     _TESTS = [{
  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_count': 2,
  90     }, {
  91         'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
  92         'info_dict': {
  93             'id': 'rivertonprisen-til-karin-fossum-1.12266449',
  94             'title': 'Rivertonprisen til Karin Fossum',
  95             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
  96         },
  97         'playlist_count': 5,
  98     }]
  99
 100     def _real_extract(self, url):
 101         playlist_id = self._match_id(url)
 102
 103         webpage = self._download_webpage(url, playlist_id)
 104
 105         entries = [
 106             self.url_result('nrk:%s' % video_id, 'NRK')
 107             for video_id in re.findall(
 108                 r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"',
 109                 webpage)
 110         ]
 111
 112         playlist_title = self._og_search_title(webpage)
 113         playlist_description = self._og_search_description(webpage)
 114
 115         return self.playlist_result(
 116             entries, playlist_id, playlist_title, playlist_description)
 117
 118
 119 class NRKTVIE(InfoExtractor):
 120     _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+))?'
 121
 122     _TESTS = [
 123         {
 124             'url': 'http://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 125             'md5': 'adf2c5454fa2bf032f47a9f8fb351342',
 126             'info_dict': {
 127                 'id': 'MUHH48000314',
 128                 'ext': 'flv',
 129                 'title': '20 spørsmål',
 130                 'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 131                 'upload_date': '20140523',
 132                 'duration': 1741.52,
 133             },
 134         },
 135         {
 136             'url': 'http://tv.nrk.no/program/mdfp15000514',
 137             'md5': '383650ece2b25ecec996ad7b5bb2a384',
 138             'info_dict': {
 139                 'id': 'mdfp15000514',
 140                 'ext': 'flv',
 141                 'title': 'Kunnskapskanalen: Grunnlovsjubiléet - Stor ståhei for ingenting',
 142                 'description': 'md5:654c12511f035aed1e42bdf5db3b206a',
 143                 'upload_date': '20140524',
 144                 'duration': 4605.0,
 145             },
 146         },
 147         {
 148             # single playlist video
 149             'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 150             'md5': 'adbd1dbd813edaf532b0a253780719c2',
 151             'info_dict': {
 152                 'id': 'MSPO40010515-part2',
 153                 'ext': 'flv',
 154                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 155                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 156                 'upload_date': '20150106',
 157             },
 158             'skip': 'Only works from Norway',
 159         },
 160         {
 161             'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 162             'playlist': [
 163                 {
 164                     'md5': '9480285eff92d64f06e02a5367970a7a',
 165                     'info_dict': {
 166                         'id': 'MSPO40010515-part1',
 167                         'ext': 'flv',
 168                         'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
 169                         'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 170                         'upload_date': '20150106',
 171                     },
 172                 },
 173                 {
 174                     'md5': 'adbd1dbd813edaf532b0a253780719c2',
 175                     'info_dict': {
 176                         'id': 'MSPO40010515-part2',
 177                         'ext': 'flv',
 178                         'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 179                         'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 180                         'upload_date': '20150106',
 181                     },
 182                 },
 183             ],
 184             'info_dict': {
 185                 'id': 'MSPO40010515',
 186                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
 187                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 188                 'upload_date': '20150106',
 189                 'duration': 6947.5199999999995,
 190             },
 191             'skip': 'Only works from Norway',
 192         }
 193     ]
 194
 195     def _debug_print(self, txt):
 196         if self._downloader.params.get('verbose', False):
 197             self.to_screen('[debug] %s' % txt)
 198
 199     def _get_subtitles(self, subtitlesurl, video_id, baseurl):
 200         url = "%s%s" % (baseurl, subtitlesurl)
 201         self._debug_print('%s: Subtitle url: %s' % (video_id, url))
 202         captions = self._download_xml(
 203             url, video_id, 'Downloading subtitles',
 204             transform_source=lambda s: s.replace(r'<br />', '\r\n'))
 205         lang = captions.get('lang', 'no')
 206         ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/ns/ttml}'))
 207         srt = ''
 208         for pos, p in enumerate(ps):
 209             begin = parse_duration(p.get('begin'))
 210             duration = parse_duration(p.get('dur'))
 211             starttime = self._subtitles_timecode(begin)
 212             endtime = self._subtitles_timecode(begin + duration)
 213             srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (compat_str(pos), starttime, endtime, p.text)
 214         return {lang: [
 215             {'ext': 'ttml', 'url': url},
 216             {'ext': 'srt', 'data': srt},
 217         ]}
 218
 219     def _extract_f4m(self, manifest_url, video_id):
 220         return self._extract_f4m_formats(manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id)
 221
 222     def _real_extract(self, url):
 223         mobj = re.match(self._VALID_URL, url)
 224         video_id = mobj.group('id')
 225         part_id = mobj.group('part_id')
 226         baseurl = mobj.group('baseurl')
 227
 228         webpage = self._download_webpage(url, video_id)
 229
 230         title = self._html_search_meta(
 231             'title', webpage, 'title')
 232         description = self._html_search_meta(
 233             'description', webpage, 'description')
 234
 235         thumbnail = self._html_search_regex(
 236             r'data-posterimage="([^"]+)"',
 237             webpage, 'thumbnail', fatal=False)
 238         upload_date = unified_strdate(self._html_search_meta(
 239             'rightsfrom', webpage, 'upload date', fatal=False))
 240         duration = float_or_none(self._html_search_regex(
 241             r'data-duration="([^"]+)"',
 242             webpage, 'duration', fatal=False))
 243
 244         # playlist
 245         parts = re.findall(
 246             r'<a href="#del=(\d+)"[^>]+data-argument="([^"]+)">([^<]+)</a>', webpage)
 247         if parts:
 248             entries = []
 249             for current_part_id, stream_url, part_title in parts:
 250                 if part_id and current_part_id != part_id:
 251                     continue
 252                 video_part_id = '%s-part%s' % (video_id, current_part_id)
 253                 formats = self._extract_f4m(stream_url, video_part_id)
 254                 entries.append({
 255                     'id': video_part_id,
 256                     'title': part_title,
 257                     'description': description,
 258                     'thumbnail': thumbnail,
 259                     'upload_date': upload_date,
 260                     'formats': formats,
 261                 })
 262             if part_id:
 263                 if entries:
 264                     return entries[0]
 265             else:
 266                 playlist = self.playlist_result(entries, video_id, title, description)
 267                 playlist.update({
 268                     'thumbnail': thumbnail,
 269                     'upload_date': upload_date,
 270                     'duration': duration,
 271                 })
 272                 return playlist
 273
 274         formats = []
 275
 276         f4m_url = re.search(r'data-media="([^"]+)"', webpage)
 277         if f4m_url:
 278             formats.extend(self._extract_f4m(f4m_url.group(1), video_id))
 279
 280         m3u8_url = re.search(r'data-hls-media="([^"]+)"', webpage)
 281         if m3u8_url:
 282             formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4'))
 283         self._sort_formats(formats)
 284
 285         subtitles_url = self._html_search_regex(
 286             r'data-subtitlesurl[ ]*=[ ]*"([^"]+)"',
 287             webpage, 'subtitle URL', default=None)
 288         subtitles = None
 289         if subtitles_url:
 290             subtitles = self.extract_subtitles(subtitles_url, video_id, baseurl)
 291
 292         return {
 293             'id': video_id,
 294             'title': title,
 295             'description': description,
 296             'thumbnail': thumbnail,
 297             'upload_date': upload_date,
 298             'duration': duration,
 299             'formats': formats,
 300             'subtitles': subtitles,
 301         }