_ 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             }
  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             }
  39         },
  40     ]
  41
  42     def _real_extract(self, url):
  43         video_id = self._match_id(url)
  44
  45         data = self._download_json(
  46             'http://v8.psapi.nrk.no/mediaelement/%s' % video_id,
  47             video_id, 'Downloading media JSON')
  48
  49         if data['usageRights']['isGeoBlocked']:
  50             raise ExtractorError(
  51                 'NRK har ikke rettig-heter til å vise dette programmet utenfor Norge',
  52                 expected=True)
  53
  54         video_url = data['mediaUrl'] + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81'
  55
  56         images = data.get('images')
  57         if images:
  58             thumbnails = images['webImages']
  59             thumbnails.sort(key=lambda image: image['pixelWidth'])
  60             thumbnail = thumbnails[-1]['imageUrl']
  61         else:
  62             thumbnail = None
  63
  64         return {
  65             'id': video_id,
  66             'url': video_url,
  67             'ext': 'flv',
  68             'title': data['title'],
  69             'description': data['description'],
  70             'thumbnail': thumbnail,
  71         }
  72
  73
  74 class NRKTVIE(InfoExtractor):
  75     _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+))?'
  76
  77     _TESTS = [
  78         {
  79             'url': 'http://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
  80             'md5': 'adf2c5454fa2bf032f47a9f8fb351342',
  81             'info_dict': {
  82                 'id': 'MUHH48000314',
  83                 'ext': 'flv',
  84                 'title': '20 spørsmål',
  85                 'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
  86                 'upload_date': '20140523',
  87                 'duration': 1741.52,
  88             },
  89         },
  90         {
  91             'url': 'http://tv.nrk.no/program/mdfp15000514',
  92             'md5': '383650ece2b25ecec996ad7b5bb2a384',
  93             'info_dict': {
  94                 'id': 'mdfp15000514',
  95                 'ext': 'flv',
  96                 'title': 'Kunnskapskanalen: Grunnlovsjubiléet - Stor ståhei for ingenting',
  97                 'description': 'md5:654c12511f035aed1e42bdf5db3b206a',
  98                 'upload_date': '20140524',
  99                 'duration': 4605.0,
 100             },
 101         },
 102         {
 103             # single playlist video
 104             'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 105             'md5': 'adbd1dbd813edaf532b0a253780719c2',
 106             'info_dict': {
 107                 'id': 'MSPO40010515-part2',
 108                 'ext': 'flv',
 109                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 110                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 111                 'upload_date': '20150106',
 112             },
 113             'skip': 'Only works from Norway',
 114         },
 115         {
 116             'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 117             'playlist': [
 118                 {
 119                     'md5': '9480285eff92d64f06e02a5367970a7a',
 120                     'info_dict': {
 121                         'id': 'MSPO40010515-part1',
 122                         'ext': 'flv',
 123                         'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
 124                         'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 125                         'upload_date': '20150106',
 126                     },
 127                 },
 128                 {
 129                     'md5': 'adbd1dbd813edaf532b0a253780719c2',
 130                     'info_dict': {
 131                         'id': 'MSPO40010515-part2',
 132                         'ext': 'flv',
 133                         'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 134                         'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 135                         'upload_date': '20150106',
 136                     },
 137                 },
 138             ],
 139             'info_dict': {
 140                 'id': 'MSPO40010515',
 141                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
 142                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 143                 'upload_date': '20150106',
 144                 'duration': 6947.5199999999995,
 145             },
 146             'skip': 'Only works from Norway',
 147         }
 148     ]
 149
 150     def _debug_print(self, txt):
 151         if self._downloader.params.get('verbose', False):
 152             self.to_screen('[debug] %s' % txt)
 153
 154     def _get_subtitles(self, subtitlesurl, video_id, baseurl):
 155         url = "%s%s" % (baseurl, subtitlesurl)
 156         self._debug_print('%s: Subtitle url: %s' % (video_id, url))
 157         captions = self._download_xml(
 158             url, video_id, 'Downloading subtitles',
 159             transform_source=lambda s: s.replace(r'<br />', '\r\n'))
 160         lang = captions.get('lang', 'no')
 161         ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/ns/ttml}'))
 162         srt = ''
 163         for pos, p in enumerate(ps):
 164             begin = parse_duration(p.get('begin'))
 165             duration = parse_duration(p.get('dur'))
 166             starttime = self._subtitles_timecode(begin)
 167             endtime = self._subtitles_timecode(begin + duration)
 168             srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (compat_str(pos), starttime, endtime, p.text)
 169         return {lang: [
 170             {'ext': 'ttml', 'url': url},
 171             {'ext': 'srt', 'data': srt},
 172         ]}
 173
 174     def _extract_f4m(self, manifest_url, video_id):
 175         return self._extract_f4m_formats(manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id)
 176
 177     def _real_extract(self, url):
 178         mobj = re.match(self._VALID_URL, url)
 179         video_id = mobj.group('id')
 180         part_id = mobj.group('part_id')
 181         baseurl = mobj.group('baseurl')
 182
 183         webpage = self._download_webpage(url, video_id)
 184
 185         title = self._html_search_meta(
 186             'title', webpage, 'title')
 187         description = self._html_search_meta(
 188             'description', webpage, 'description')
 189
 190         thumbnail = self._html_search_regex(
 191             r'data-posterimage="([^"]+)"',
 192             webpage, 'thumbnail', fatal=False)
 193         upload_date = unified_strdate(self._html_search_meta(
 194             'rightsfrom', webpage, 'upload date', fatal=False))
 195         duration = float_or_none(self._html_search_regex(
 196             r'data-duration="([^"]+)"',
 197             webpage, 'duration', fatal=False))
 198
 199         # playlist
 200         parts = re.findall(
 201             r'<a href="#del=(\d+)"[^>]+data-argument="([^"]+)">([^<]+)</a>', webpage)
 202         if parts:
 203             entries = []
 204             for current_part_id, stream_url, part_title in parts:
 205                 if part_id and current_part_id != part_id:
 206                     continue
 207                 video_part_id = '%s-part%s' % (video_id, current_part_id)
 208                 formats = self._extract_f4m(stream_url, video_part_id)
 209                 entries.append({
 210                     'id': video_part_id,
 211                     'title': part_title,
 212                     'description': description,
 213                     'thumbnail': thumbnail,
 214                     'upload_date': upload_date,
 215                     'formats': formats,
 216                 })
 217             if part_id:
 218                 if entries:
 219                     return entries[0]
 220             else:
 221                 playlist = self.playlist_result(entries, video_id, title, description)
 222                 playlist.update({
 223                     'thumbnail': thumbnail,
 224                     'upload_date': upload_date,
 225                     'duration': duration,
 226                 })
 227                 return playlist
 228
 229         formats = []
 230
 231         f4m_url = re.search(r'data-media="([^"]+)"', webpage)
 232         if f4m_url:
 233             formats.extend(self._extract_f4m(f4m_url.group(1), video_id))
 234
 235         m3u8_url = re.search(r'data-hls-media="([^"]+)"', webpage)
 236         if m3u8_url:
 237             formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4'))
 238         self._sort_formats(formats)
 239
 240         subtitles_url = self._html_search_regex(
 241             r'data-subtitlesurl[ ]*=[ ]*"([^"]+)"',
 242             webpage, 'subtitle URL', default=None)
 243         subtitles = None
 244         if subtitles_url:
 245             subtitles = self.extract_subtitles(subtitles_url, video_id, baseurl)
 246
 247         return {
 248             'id': video_id,
 249             'title': title,
 250             'description': description,
 251             'thumbnail': thumbnail,
 252             'upload_date': upload_date,
 253             'duration': duration,
 254             'formats': formats,
 255             'subtitles': subtitles,
 256         }