yt_dlp/extractor/eagleplatform.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_HTTPError
   5 from ..utils import (
   6     ExtractorError,
   7     int_or_none,
   8     unsmuggle_url,
   9     url_or_none,
  10 )
  11
  12
  13 class EaglePlatformIE(InfoExtractor):
  14     _VALID_URL = r'''(?x)
  15                     (?:
  16                         eagleplatform:(?P<custom_host>[^/]+):|
  17                         https?://(?P<host>.+?\.media\.eagleplatform\.com)/index/player\?.*\brecord_id=
  18                     )
  19                     (?P<id>\d+)
  20                 '''
  21     _TESTS = [{
  22         # http://lenta.ru/news/2015/03/06/navalny/
  23         'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',
  24         # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
  25         'info_dict': {
  26             'id': '227304',
  27             'ext': 'mp4',
  28             'title': 'Навальный вышел на свободу',
  29             'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
  30             'thumbnail': r're:^https?://.*\.jpg$',
  31             'duration': 87,
  32             'view_count': int,
  33             'age_limit': 0,
  34         },
  35     }, {
  36         # http://muz-tv.ru/play/7129/
  37         # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true
  38         'url': 'eagleplatform:media.clipyou.ru:12820',
  39         'md5': '358597369cf8ba56675c1df15e7af624',
  40         'info_dict': {
  41             'id': '12820',
  42             'ext': 'mp4',
  43             'title': "'O Sole Mio",
  44             'thumbnail': r're:^https?://.*\.jpg$',
  45             'duration': 216,
  46             'view_count': int,
  47         },
  48         'skip': 'Georestricted',
  49     }, {
  50         # referrer protected video (https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/)
  51         'url': 'eagleplatform:tvrainru.media.eagleplatform.com:582306',
  52         'only_matching': True,
  53     }]
  54
  55     @staticmethod
  56     def _extract_url(webpage):
  57         # Regular iframe embedding
  58         mobj = re.search(
  59             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1',
  60             webpage)
  61         if mobj is not None:
  62             return mobj.group('url')
  63         PLAYER_JS_RE = r'''
  64                         <script[^>]+
  65                             src=(?P<qjs>["\'])(?:https?:)?//(?P<host>(?:(?!(?P=qjs)).)+\.media\.eagleplatform\.com)/player/player\.js(?P=qjs)
  66                         .+?
  67                     '''
  68         # "Basic usage" embedding (see http://dultonmedia.github.io/eplayer/)
  69         mobj = re.search(
  70             r'''(?xs)
  71                     %s
  72                     <div[^>]+
  73                         class=(?P<qclass>["\'])eagleplayer(?P=qclass)[^>]+
  74                         data-id=["\'](?P<id>\d+)
  75             ''' % PLAYER_JS_RE, webpage)
  76         if mobj is not None:
  77             return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
  78         # Generalization of "Javascript code usage", "Combined usage" and
  79         # "Usage without attaching to DOM" embeddings (see
  80         # http://dultonmedia.github.io/eplayer/)
  81         mobj = re.search(
  82             r'''(?xs)
  83                     %s
  84                     <script>
  85                     .+?
  86                     new\s+EaglePlayer\(
  87                         (?:[^,]+\s*,\s*)?
  88                         {
  89                             .+?
  90                             \bid\s*:\s*["\']?(?P<id>\d+)
  91                             .+?
  92                         }
  93                     \s*\)
  94                     .+?
  95                     </script>
  96             ''' % PLAYER_JS_RE, webpage)
  97         if mobj is not None:
  98             return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
  99
 100     @staticmethod
 101     def _handle_error(response):
 102         status = int_or_none(response.get('status', 200))
 103         if status != 200:
 104             raise ExtractorError(' '.join(response['errors']), expected=True)
 105
 106     def _download_json(self, url_or_request, video_id, *args, **kwargs):
 107         try:
 108             response = super(EaglePlatformIE, self)._download_json(
 109                 url_or_request, video_id, *args, **kwargs)
 110         except ExtractorError as ee:
 111             if isinstance(ee.cause, compat_HTTPError):
 112                 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
 113                 self._handle_error(response)
 114             raise
 115         return response
 116
 117     def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'):
 118         return self._download_json(url_or_request, video_id, note)['data'][0]
 119
 120     def _real_extract(self, url):
 121         url, smuggled_data = unsmuggle_url(url, {})
 122
 123         mobj = self._match_valid_url(url)
 124         host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
 125
 126         headers = {}
 127         query = {
 128             'id': video_id,
 129         }
 130
 131         referrer = smuggled_data.get('referrer')
 132         if referrer:
 133             headers['Referer'] = referrer
 134             query['referrer'] = referrer
 135
 136         player_data = self._download_json(
 137             'http://%s/api/player_data' % host, video_id,
 138             headers=headers, query=query)
 139
 140         media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
 141
 142         title = media['title']
 143         description = media.get('description')
 144         thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:')
 145         duration = int_or_none(media.get('duration'))
 146         view_count = int_or_none(media.get('views'))
 147
 148         age_restriction = media.get('age_restriction')
 149         age_limit = None
 150         if age_restriction:
 151             age_limit = 0 if age_restriction == 'allow_all' else 18
 152
 153         secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:')
 154
 155         formats = []
 156
 157         m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
 158         m3u8_formats = self._extract_m3u8_formats(
 159             m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
 160             m3u8_id='hls', fatal=False)
 161         formats.extend(m3u8_formats)
 162
 163         m3u8_formats_dict = {}
 164         for f in m3u8_formats:
 165             if f.get('height') is not None:
 166                 m3u8_formats_dict[f['height']] = f
 167
 168         mp4_data = self._download_json(
 169             # Secure mp4 URL is constructed according to Player.prototype.mp4 from
 170             # http://lentaru.media.eagleplatform.com/player/player.js
 171             re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4s', secure_m3u8),
 172             video_id, 'Downloading mp4 JSON', fatal=False)
 173         if mp4_data:
 174             for format_id, format_url in mp4_data.get('data', {}).items():
 175                 if not url_or_none(format_url):
 176                     continue
 177                 height = int_or_none(format_id)
 178                 if height is not None and m3u8_formats_dict.get(height):
 179                     f = m3u8_formats_dict[height].copy()
 180                     f.update({
 181                         'format_id': f['format_id'].replace('hls', 'http'),
 182                         'protocol': 'http',
 183                     })
 184                 else:
 185                     f = {
 186                         'format_id': 'http-%s' % format_id,
 187                         'height': int_or_none(format_id),
 188                     }
 189                 f['url'] = format_url
 190                 formats.append(f)
 191
 192         self._sort_formats(formats)
 193
 194         return {
 195             'id': video_id,
 196             'title': title,
 197             'description': description,
 198             'thumbnail': thumbnail,
 199             'duration': duration,
 200             'view_count': view_count,
 201             'age_limit': age_limit,
 202             'formats': formats,
 203         }