_ Git - youtube-dl/blob - youtube_dl/extractor/eagleplatform.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_HTTPError,
   9     compat_str,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14 )
  15
  16
  17 class EaglePlatformIE(InfoExtractor):
  18     _VALID_URL = r'''(?x)
  19                     (?:
  20                         eagleplatform:(?P<custom_host>[^/]+):|
  21                         https?://(?P<host>.+?\.media\.eagleplatform\.com)/index/player\?.*\brecord_id=
  22                     )
  23                     (?P<id>\d+)
  24                 '''
  25     _TESTS = [{
  26         # http://lenta.ru/news/2015/03/06/navalny/
  27         'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',
  28         # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
  29         'info_dict': {
  30             'id': '227304',
  31             'ext': 'mp4',
  32             'title': 'Навальный вышел на свободу',
  33             'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
  34             'thumbnail': r're:^https?://.*\.jpg$',
  35             'duration': 87,
  36             'view_count': int,
  37             'age_limit': 0,
  38         },
  39     }, {
  40         # http://muz-tv.ru/play/7129/
  41         # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true
  42         'url': 'eagleplatform:media.clipyou.ru:12820',
  43         'md5': '358597369cf8ba56675c1df15e7af624',
  44         'info_dict': {
  45             'id': '12820',
  46             'ext': 'mp4',
  47             'title': "'O Sole Mio",
  48             'thumbnail': r're:^https?://.*\.jpg$',
  49             'duration': 216,
  50             'view_count': int,
  51         },
  52         'skip': 'Georestricted',
  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, note='Downloading JSON metadata', *args, **kwargs):
 107         try:
 108             response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
 109         except ExtractorError as ee:
 110             if isinstance(ee.cause, compat_HTTPError):
 111                 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
 112                 self._handle_error(response)
 113             raise
 114         return response
 115
 116     def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'):
 117         return self._download_json(url_or_request, video_id, note)['data'][0]
 118
 119     def _real_extract(self, url):
 120         mobj = re.match(self._VALID_URL, url)
 121         host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
 122
 123         player_data = self._download_json(
 124             'http://%s/api/player_data?id=%s' % (host, video_id), video_id)
 125
 126         media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
 127
 128         title = media['title']
 129         description = media.get('description')
 130         thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:')
 131         duration = int_or_none(media.get('duration'))
 132         view_count = int_or_none(media.get('views'))
 133
 134         age_restriction = media.get('age_restriction')
 135         age_limit = None
 136         if age_restriction:
 137             age_limit = 0 if age_restriction == 'allow_all' else 18
 138
 139         secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:')
 140
 141         formats = []
 142
 143         m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
 144         m3u8_formats = self._extract_m3u8_formats(
 145             m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
 146             m3u8_id='hls', fatal=False)
 147         formats.extend(m3u8_formats)
 148
 149         m3u8_formats_dict = {}
 150         for f in m3u8_formats:
 151             if f.get('height') is not None:
 152                 m3u8_formats_dict[f['height']] = f
 153
 154         mp4_data = self._download_json(
 155             # Secure mp4 URL is constructed according to Player.prototype.mp4 from
 156             # http://lentaru.media.eagleplatform.com/player/player.js
 157             re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4s', secure_m3u8),
 158             video_id, 'Downloading mp4 JSON', fatal=False)
 159         if mp4_data:
 160             for format_id, format_url in mp4_data.get('data', {}).items():
 161                 if not isinstance(format_url, compat_str):
 162                     continue
 163                 height = int_or_none(format_id)
 164                 if height is not None and m3u8_formats_dict.get(height):
 165                     f = m3u8_formats_dict[height].copy()
 166                     f.update({
 167                         'format_id': f['format_id'].replace('hls', 'http'),
 168                         'protocol': 'http',
 169                     })
 170                 else:
 171                     f = {
 172                         'format_id': 'http-%s' % format_id,
 173                         'height': int_or_none(format_id),
 174                     }
 175                 f['url'] = format_url
 176                 formats.append(f)
 177
 178         self._sort_formats(formats)
 179
 180         return {
 181             'id': video_id,
 182             'title': title,
 183             'description': description,
 184             'thumbnail': thumbnail,
 185             'duration': duration,
 186             'view_count': view_count,
 187             'age_limit': age_limit,
 188             'formats': formats,
 189         }