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 ..utils import (
   8     ExtractorError,
   9     float_or_none,
  10     parse_duration,
  11     unified_strdate,
  12 )
  13
  14
  15 class NRKIE(InfoExtractor):
  16     _VALID_URL = r'(?:nrk:|http://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
  17
  18     _TESTS = [
  19         {
  20             'url': 'http://www.nrk.no/video/PS*150533',
  21             'md5': 'bccd850baebefe23b56d708a113229c2',
  22             'info_dict': {
  23                 'id': '150533',
  24                 'ext': 'flv',
  25                 'title': 'Dompap og andre fugler i Piip-Show',
  26                 'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
  27                 'duration': 263,
  28             }
  29         },
  30         {
  31             'url': 'http://www.nrk.no/video/PS*154915',
  32             'md5': '0b1493ba1aae7d9579a5ad5531bc395a',
  33             'info_dict': {
  34                 'id': '154915',
  35                 'ext': 'flv',
  36                 'title': 'Slik høres internett ut når du er blind',
  37                 'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
  38                 'duration': 20,
  39             }
  40         },
  41     ]
  42
  43     def _real_extract(self, url):
  44         video_id = self._match_id(url)
  45
  46         data = self._download_json(
  47             'http://v8.psapi.nrk.no/mediaelement/%s' % video_id,
  48             video_id, 'Downloading media JSON')
  49
  50         if data['usageRights']['isGeoBlocked']:
  51             raise ExtractorError(
  52                 'NRK har ikke rettig-heter til å vise dette programmet utenfor Norge',
  53                 expected=True)
  54
  55         video_url = data['mediaUrl'] + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81'
  56
  57         duration = parse_duration(data.get('duration'))
  58
  59         images = data.get('images')
  60         if images:
  61             thumbnails = images['webImages']
  62             thumbnails.sort(key=lambda image: image['pixelWidth'])
  63             thumbnail = thumbnails[-1]['imageUrl']
  64         else:
  65             thumbnail = None
  66
  67         return {
  68             'id': video_id,
  69             'url': video_url,
  70             'ext': 'flv',
  71             'title': data['title'],
  72             'description': data['description'],
  73             'duration': duration,
  74             'thumbnail': thumbnail,
  75         }
  76
  77
  78 class NRKPlaylistIE(InfoExtractor):
  79     _VALID_URL = r'http://(?:www\.)?nrk\.no/(?!video)(?:[^/]+/)+(?P<id>[^/]+)'
  80
  81     _TESTS = [{
  82         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
  83         'info_dict': {
  84             'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
  85             'title': 'Gjenopplev den historiske solformørkelsen',
  86             'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
  87         },
  88         'playlist_count': 2,
  89     }, {
  90         'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
  91         'info_dict': {
  92             'id': 'rivertonprisen-til-karin-fossum-1.12266449',
  93             'title': 'Rivertonprisen til Karin Fossum',
  94             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
  95         },
  96         'playlist_count': 5,
  97     }]
  98
  99     def _real_extract(self, url):
 100         playlist_id = self._match_id(url)
 101
 102         webpage = self._download_webpage(url, playlist_id)
 103
 104         entries = [
 105             self.url_result('nrk:%s' % video_id, 'NRK')
 106             for video_id in re.findall(
 107                 r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"',
 108                 webpage)
 109         ]
 110
 111         playlist_title = self._og_search_title(webpage)
 112         playlist_description = self._og_search_description(webpage)
 113
 114         return self.playlist_result(
 115             entries, playlist_id, playlist_title, playlist_description)
 116
 117
 118 class NRKTVIE(InfoExtractor):
 119     _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+))?'
 120
 121     _TESTS = [
 122         {
 123             'url': 'http://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 124             'md5': 'adf2c5454fa2bf032f47a9f8fb351342',
 125             'info_dict': {
 126                 'id': 'MUHH48000314',
 127                 'ext': 'flv',
 128                 'title': '20 spørsmål',
 129                 'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 130                 'upload_date': '20140523',
 131                 'duration': 1741.52,
 132             },
 133         },
 134         {
 135             'url': 'http://tv.nrk.no/program/mdfp15000514',
 136             'md5': '383650ece2b25ecec996ad7b5bb2a384',
 137             'info_dict': {
 138                 'id': 'mdfp15000514',
 139                 'ext': 'flv',
 140                 'title': 'Kunnskapskanalen: Grunnlovsjubiléet - Stor ståhei for ingenting',
 141                 'description': 'md5:654c12511f035aed1e42bdf5db3b206a',
 142                 'upload_date': '20140524',
 143                 'duration': 4605.0,
 144             },
 145         },
 146         {
 147             # single playlist video
 148             'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 149             'md5': 'adbd1dbd813edaf532b0a253780719c2',
 150             'info_dict': {
 151                 'id': 'MSPO40010515-part2',
 152                 'ext': 'flv',
 153                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 154                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 155                 'upload_date': '20150106',
 156             },
 157             'skip': 'Only works from Norway',
 158         },
 159         {
 160             'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 161             'playlist': [
 162                 {
 163                     'md5': '9480285eff92d64f06e02a5367970a7a',
 164                     'info_dict': {
 165                         'id': 'MSPO40010515-part1',
 166                         'ext': 'flv',
 167                         'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
 168                         'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 169                         'upload_date': '20150106',
 170                     },
 171                 },
 172                 {
 173                     'md5': 'adbd1dbd813edaf532b0a253780719c2',
 174                     'info_dict': {
 175                         'id': 'MSPO40010515-part2',
 176                         'ext': 'flv',
 177                         'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 178                         'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 179                         'upload_date': '20150106',
 180                     },
 181                 },
 182             ],
 183             'info_dict': {
 184                 'id': 'MSPO40010515',
 185                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
 186                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 187                 'upload_date': '20150106',
 188                 'duration': 6947.5199999999995,
 189             },
 190             'skip': 'Only works from Norway',
 191         }
 192     ]
 193
 194     def _debug_print(self, txt):
 195         if self._downloader.params.get('verbose', False):
 196             self.to_screen('[debug] %s' % txt)
 197
 198     def _get_subtitles(self, subtitlesurl, video_id, baseurl):
 199         url = "%s%s" % (baseurl, subtitlesurl)
 200         self._debug_print('%s: Subtitle url: %s' % (video_id, url))
 201         captions = self._download_xml(
 202             url, video_id, 'Downloading subtitles')
 203         lang = captions.get('lang', 'no')
 204         return {lang: [
 205             {'ext': 'ttml', 'url': url},
 206         ]}
 207
 208     def _extract_f4m(self, manifest_url, video_id):
 209         return self._extract_f4m_formats(manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id)
 210
 211     def _real_extract(self, url):
 212         mobj = re.match(self._VALID_URL, url)
 213         video_id = mobj.group('id')
 214         part_id = mobj.group('part_id')
 215         baseurl = mobj.group('baseurl')
 216
 217         webpage = self._download_webpage(url, video_id)
 218
 219         title = self._html_search_meta(
 220             'title', webpage, 'title')
 221         description = self._html_search_meta(
 222             'description', webpage, 'description')
 223
 224         thumbnail = self._html_search_regex(
 225             r'data-posterimage="([^"]+)"',
 226             webpage, 'thumbnail', fatal=False)
 227         upload_date = unified_strdate(self._html_search_meta(
 228             'rightsfrom', webpage, 'upload date', fatal=False))
 229         duration = float_or_none(self._html_search_regex(
 230             r'data-duration="([^"]+)"',
 231             webpage, 'duration', fatal=False))
 232
 233         # playlist
 234         parts = re.findall(
 235             r'<a href="#del=(\d+)"[^>]+data-argument="([^"]+)">([^<]+)</a>', webpage)
 236         if parts:
 237             entries = []
 238             for current_part_id, stream_url, part_title in parts:
 239                 if part_id and current_part_id != part_id:
 240                     continue
 241                 video_part_id = '%s-part%s' % (video_id, current_part_id)
 242                 formats = self._extract_f4m(stream_url, video_part_id)
 243                 entries.append({
 244                     'id': video_part_id,
 245                     'title': part_title,
 246                     'description': description,
 247                     'thumbnail': thumbnail,
 248                     'upload_date': upload_date,
 249                     'formats': formats,
 250                 })
 251             if part_id:
 252                 if entries:
 253                     return entries[0]
 254             else:
 255                 playlist = self.playlist_result(entries, video_id, title, description)
 256                 playlist.update({
 257                     'thumbnail': thumbnail,
 258                     'upload_date': upload_date,
 259                     'duration': duration,
 260                 })
 261                 return playlist
 262
 263         formats = []
 264
 265         f4m_url = re.search(r'data-media="([^"]+)"', webpage)
 266         if f4m_url:
 267             formats.extend(self._extract_f4m(f4m_url.group(1), video_id))
 268
 269         m3u8_url = re.search(r'data-hls-media="([^"]+)"', webpage)
 270         if m3u8_url:
 271             formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4'))
 272         self._sort_formats(formats)
 273
 274         subtitles_url = self._html_search_regex(
 275             r'data-subtitlesurl[ ]*=[ ]*"([^"]+)"',
 276             webpage, 'subtitle URL', default=None)
 277         subtitles = None
 278         if subtitles_url:
 279             subtitles = self.extract_subtitles(subtitles_url, video_id, baseurl)
 280
 281         return {
 282             'id': video_id,
 283             'title': title,
 284             'description': description,
 285             'thumbnail': thumbnail,
 286             'upload_date': upload_date,
 287             'duration': duration,
 288             'formats': formats,
 289             'subtitles': subtitles,
 290         }