Raphaël G. Git Repositories - youtubedl/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_urllib_parse_unquote
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     parse_age_limit,
  12     parse_duration,
  13 )
  14
  15
  16 class NRKBaseIE(InfoExtractor):
  17     def _extract_formats(self, manifest_url, video_id, fatal=True):
  18         formats = []
  19         formats.extend(self._extract_f4m_formats(
  20             manifest_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81',
  21             video_id, f4m_id='hds', fatal=fatal))
  22         formats.extend(self._extract_m3u8_formats(manifest_url.replace(
  23             'akamaihd.net/z/', 'akamaihd.net/i/').replace('/manifest.f4m', '/master.m3u8'),
  24             video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=fatal))
  25         return formats
  26
  27     def _real_extract(self, url):
  28         video_id = self._match_id(url)
  29
  30         data = self._download_json(
  31             'http://%s/mediaelement/%s' % (self._API_HOST, video_id),
  32             video_id, 'Downloading mediaelement JSON')
  33
  34         title = data.get('fullTitle') or data.get('mainTitle') or data['title']
  35         video_id = data.get('id') or video_id
  36
  37         entries = []
  38
  39         media_assets = data.get('mediaAssets')
  40         if media_assets and isinstance(media_assets, list):
  41             def video_id_and_title(idx):
  42                 return ((video_id, title) if len(media_assets) == 1
  43                         else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
  44             for num, asset in enumerate(media_assets, 1):
  45                 asset_url = asset.get('url')
  46                 if not asset_url:
  47                     continue
  48                 formats = self._extract_formats(asset_url, video_id, fatal=False)
  49                 if not formats:
  50                     continue
  51                 self._sort_formats(formats)
  52                 entry_id, entry_title = video_id_and_title(num)
  53                 duration = parse_duration(asset.get('duration'))
  54                 subtitles = {}
  55                 for subtitle in ('webVtt', 'timedText'):
  56                     subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
  57                     if subtitle_url:
  58                         subtitles.setdefault('no', []).append({
  59                             'url': compat_urllib_parse_unquote(subtitle_url)
  60                         })
  61                 entries.append({
  62                     'id': asset.get('carrierId') or entry_id,
  63                     'title': entry_title,
  64                     'duration': duration,
  65                     'subtitles': subtitles,
  66                     'formats': formats,
  67                 })
  68
  69         if not entries:
  70             media_url = data.get('mediaUrl')
  71             if media_url:
  72                 formats = self._extract_formats(media_url, video_id)
  73                 self._sort_formats(formats)
  74                 duration = parse_duration(data.get('duration'))
  75                 entries = [{
  76                     'id': video_id,
  77                     'title': title,
  78                     'duration': duration,
  79                     'formats': formats,
  80                 }]
  81
  82         if not entries:
  83             if data.get('usageRights', {}).get('isGeoBlocked'):
  84                 raise ExtractorError(
  85                     'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
  86                     expected=True)
  87
  88         conviva = data.get('convivaStatistics') or {}
  89         series = conviva.get('seriesName') or data.get('seriesTitle')
  90         episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
  91
  92         thumbnails = None
  93         images = data.get('images')
  94         if images and isinstance(images, dict):
  95             web_images = images.get('webImages')
  96             if isinstance(web_images, list):
  97                 thumbnails = [{
  98                     'url': image['imageUrl'],
  99                     'width': int_or_none(image.get('width')),
 100                     'height': int_or_none(image.get('height')),
 101                 } for image in web_images if image.get('imageUrl')]
 102
 103         description = data.get('description')
 104
 105         common_info = {
 106             'description': description,
 107             'series': series,
 108             'episode': episode,
 109             'age_limit': parse_age_limit(data.get('legalAge')),
 110             'thumbnails': thumbnails,
 111         }
 112
 113         vcodec = 'none' if data.get('mediaType') == 'Audio' else None
 114
 115         # TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged
 116
 117         for entry in entries:
 118             entry.update(common_info)
 119             for f in entry['formats']:
 120                 f['vcodec'] = vcodec
 121
 122         return self.playlist_result(entries, video_id, title, description)
 123
 124
 125 class NRKIE(NRKBaseIE):
 126     _VALID_URL = r'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
 127     _API_HOST = 'v8.psapi.nrk.no'
 128     _TESTS = [{
 129         # video
 130         'url': 'http://www.nrk.no/video/PS*150533',
 131         'md5': '2f7f6eeb2aacdd99885f355428715cfa',
 132         'info_dict': {
 133             'id': '150533',
 134             'ext': 'mp4',
 135             'title': 'Dompap og andre fugler i Piip-Show',
 136             'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
 137             'duration': 263,
 138         }
 139     }, {
 140         # audio
 141         'url': 'http://www.nrk.no/video/PS*154915',
 142         # MD5 is unstable
 143         'info_dict': {
 144             'id': '154915',
 145             'ext': 'flv',
 146             'title': 'Slik høres internett ut når du er blind',
 147             'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
 148             'duration': 20,
 149         }
 150     }]
 151
 152
 153 class NRKTVIE(NRKBaseIE):
 154     IE_DESC = 'NRK TV and NRK Radio'
 155     _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
 156     _API_HOST = 'psapi-we.nrk.no'
 157
 158     _TESTS = [{
 159         'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 160         'md5': '4e9ca6629f09e588ed240fb11619922a',
 161         'info_dict': {
 162             'id': 'MUHH48000314AA',
 163             'ext': 'mp4',
 164             'title': '20 spørsmål 23.05.2014',
 165             'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 166             'duration': 1741,
 167         },
 168     }, {
 169         'url': 'https://tv.nrk.no/program/mdfp15000514',
 170         'md5': '43d0be26663d380603a9cf0c24366531',
 171         'info_dict': {
 172             'id': 'MDFP15000514CA',
 173             'ext': 'mp4',
 174             'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
 175             'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
 176             'duration': 4605,
 177         },
 178     }, {
 179         # single playlist video
 180         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 181         'md5': 'adbd1dbd813edaf532b0a253780719c2',
 182         'info_dict': {
 183             'id': 'MSPO40010515-part2',
 184             'ext': 'flv',
 185             'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 186             'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 187         },
 188         'skip': 'Only works from Norway',
 189     }, {
 190         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 191         'playlist': [{
 192             'md5': '9480285eff92d64f06e02a5367970a7a',
 193             'info_dict': {
 194                 'id': 'MSPO40010515-part1',
 195                 'ext': 'flv',
 196                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
 197                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 198             },
 199         }, {
 200             'md5': 'adbd1dbd813edaf532b0a253780719c2',
 201             'info_dict': {
 202                 'id': 'MSPO40010515-part2',
 203                 'ext': 'flv',
 204                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 205                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 206             },
 207         }],
 208         'info_dict': {
 209             'id': 'MSPO40010515',
 210             'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
 211             'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 212             'duration': 6947.52,
 213         },
 214         'skip': 'Only works from Norway',
 215     }, {
 216         'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
 217         'only_matching': True,
 218     }]
 219
 220
 221 class NRKPlaylistIE(InfoExtractor):
 222     _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
 223
 224     _TESTS = [{
 225         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
 226         'info_dict': {
 227             'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
 228             'title': 'Gjenopplev den historiske solformørkelsen',
 229             'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
 230         },
 231         'playlist_count': 2,
 232     }, {
 233         'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
 234         'info_dict': {
 235             'id': 'rivertonprisen-til-karin-fossum-1.12266449',
 236             'title': 'Rivertonprisen til Karin Fossum',
 237             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
 238         },
 239         'playlist_count': 5,
 240     }]
 241
 242     def _real_extract(self, url):
 243         playlist_id = self._match_id(url)
 244
 245         webpage = self._download_webpage(url, playlist_id)
 246
 247         entries = [
 248             self.url_result('nrk:%s' % video_id, 'NRK')
 249             for video_id in re.findall(
 250                 r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"',
 251                 webpage)
 252         ]
 253
 254         playlist_title = self._og_search_title(webpage)
 255         playlist_description = self._og_search_description(webpage)
 256
 257         return self.playlist_result(
 258             entries, playlist_id, playlist_title, playlist_description)
 259
 260
 261 class NRKSkoleIE(InfoExtractor):
 262     IE_DESC = 'NRK Skole'
 263     _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
 264
 265     _TESTS = [{
 266         'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
 267         'md5': '6bc936b01f9dd8ed45bc58b252b2d9b6',
 268         'info_dict': {
 269             'id': '6021',
 270             'ext': 'mp4',
 271             'title': 'Genetikk og eneggede tvillinger',
 272             'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
 273             'duration': 399,
 274         },
 275     }, {
 276         'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
 277         'only_matching': True,
 278     }]
 279
 280     def _real_extract(self, url):
 281         video_id = self._match_id(url)
 282
 283         webpage = self._download_webpage(
 284             'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
 285             video_id)
 286
 287         nrk_id = self._parse_json(
 288             self._search_regex(
 289                 r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
 290                 webpage, 'application json'),
 291             video_id)['activeMedia']['psId']
 292
 293         return self.url_result('nrk:%s' % nrk_id)