Raphaël G. Git Repositories - youtubedl/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     unsmuggle_url,
  15 )
  16
  17
  18 class EaglePlatformIE(InfoExtractor):
  19     _VALID_URL = r'''(?x)
  20                     (?:
  21                         eagleplatform:(?P<custom_host>[^/]+):|
  22                         https?://(?P<host>.+?\.media\.eagleplatform\.com)/index/player\?.*\brecord_id=
  23                     )
  24                     (?P<id>\d+)
  25                 '''
  26     _TESTS = [{
  27         # http://lenta.ru/news/2015/03/06/navalny/
  28         'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',
  29         # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
  30         'info_dict': {
  31             'id': '227304',
  32             'ext': 'mp4',
  33             'title': 'Навальный вышел на свободу',
  34             'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
  35             'thumbnail': r're:^https?://.*\.jpg$',
  36             'duration': 87,
  37             'view_count': int,
  38             'age_limit': 0,
  39         },
  40     }, {
  41         # http://muz-tv.ru/play/7129/
  42         # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true
  43         'url': 'eagleplatform:media.clipyou.ru:12820',
  44         'md5': '358597369cf8ba56675c1df15e7af624',
  45         'info_dict': {
  46             'id': '12820',
  47             'ext': 'mp4',
  48             'title': "'O Sole Mio",
  49             'thumbnail': r're:^https?://.*\.jpg$',
  50             'duration': 216,
  51             'view_count': int,
  52         },
  53         'skip': 'Georestricted',
  54     }, {
  55         # referrer protected video (https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/)
  56         'url': 'eagleplatform:tvrainru.media.eagleplatform.com:582306',
  57         'only_matching': True,
  58     }]
  59
  60     @staticmethod
  61     def _extract_url(webpage):
  62         # Regular iframe embedding
  63         mobj = re.search(
  64             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1',
  65             webpage)
  66         if mobj is not None:
  67             return mobj.group('url')
  68         PLAYER_JS_RE = r'''
  69                         <script[^>]+
  70                             src=(?P<qjs>["\'])(?:https?:)?//(?P<host>(?:(?!(?P=qjs)).)+\.media\.eagleplatform\.com)/player/player\.js(?P=qjs)
  71                         .+?
  72                     '''
  73         # "Basic usage" embedding (see http://dultonmedia.github.io/eplayer/)
  74         mobj = re.search(
  75             r'''(?xs)
  76                     %s
  77                     <div[^>]+
  78                         class=(?P<qclass>["\'])eagleplayer(?P=qclass)[^>]+
  79                         data-id=["\'](?P<id>\d+)
  80             ''' % PLAYER_JS_RE, webpage)
  81         if mobj is not None:
  82             return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
  83         # Generalization of "Javascript code usage", "Combined usage" and
  84         # "Usage without attaching to DOM" embeddings (see
  85         # http://dultonmedia.github.io/eplayer/)
  86         mobj = re.search(
  87             r'''(?xs)
  88                     %s
  89                     <script>
  90                     .+?
  91                     new\s+EaglePlayer\(
  92                         (?:[^,]+\s*,\s*)?
  93                         {
  94                             .+?
  95                             \bid\s*:\s*["\']?(?P<id>\d+)
  96                             .+?
  97                         }
  98                     \s*\)
  99                     .+?
 100                     </script>
 101             ''' % PLAYER_JS_RE, webpage)
 102         if mobj is not None:
 103             return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
 104
 105     @staticmethod
 106     def _handle_error(response):
 107         status = int_or_none(response.get('status', 200))
 108         if status != 200:
 109             raise ExtractorError(' '.join(response['errors']), expected=True)
 110
 111     def _download_json(self, url_or_request, video_id, *args, **kwargs):
 112         try:
 113             response = super(EaglePlatformIE, self)._download_json(
 114                 url_or_request, video_id, *args, **kwargs)
 115         except ExtractorError as ee:
 116             if isinstance(ee.cause, compat_HTTPError):
 117                 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
 118                 self._handle_error(response)
 119             raise
 120         return response
 121
 122     def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'):
 123         return self._download_json(url_or_request, video_id, note)['data'][0]
 124
 125     def _real_extract(self, url):
 126         url, smuggled_data = unsmuggle_url(url, {})
 127
 128         mobj = re.match(self._VALID_URL, url)
 129         host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
 130
 131         headers = {}
 132         query = {
 133             'id': video_id,
 134         }
 135
 136         referrer = smuggled_data.get('referrer')
 137         if referrer:
 138             headers['Referer'] = referrer
 139             query['referrer'] = referrer
 140
 141         player_data = self._download_json(
 142             'http://%s/api/player_data' % host, video_id,
 143             headers=headers, query=query)
 144
 145         media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
 146
 147         title = media['title']
 148         description = media.get('description')
 149         thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:')
 150         duration = int_or_none(media.get('duration'))
 151         view_count = int_or_none(media.get('views'))
 152
 153         age_restriction = media.get('age_restriction')
 154         age_limit = None
 155         if age_restriction:
 156             age_limit = 0 if age_restriction == 'allow_all' else 18
 157
 158         secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:')
 159
 160         formats = []
 161
 162         m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
 163         m3u8_formats = self._extract_m3u8_formats(
 164             m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
 165             m3u8_id='hls', fatal=False)
 166         formats.extend(m3u8_formats)
 167
 168         m3u8_formats_dict = {}
 169         for f in m3u8_formats:
 170             if f.get('height') is not None:
 171                 m3u8_formats_dict[f['height']] = f
 172
 173         mp4_data = self._download_json(
 174             # Secure mp4 URL is constructed according to Player.prototype.mp4 from
 175             # http://lentaru.media.eagleplatform.com/player/player.js
 176             re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4s', secure_m3u8),
 177             video_id, 'Downloading mp4 JSON', fatal=False)
 178         if mp4_data:
 179             for format_id, format_url in mp4_data.get('data', {}).items():
 180                 if not isinstance(format_url, compat_str):
 181                     continue
 182                 height = int_or_none(format_id)
 183                 if height is not None and m3u8_formats_dict.get(height):
 184                     f = m3u8_formats_dict[height].copy()
 185                     f.update({
 186                         'format_id': f['format_id'].replace('hls', 'http'),
 187                         'protocol': 'http',
 188                     })
 189                 else:
 190                     f = {
 191                         'format_id': 'http-%s' % format_id,
 192                         'height': int_or_none(format_id),
 193                     }
 194                 f['url'] = format_url
 195                 formats.append(f)
 196
 197         self._sort_formats(formats)
 198
 199         return {
 200             'id': video_id,
 201             'title': title,
 202             'description': description,
 203             'thumbnail': thumbnail,
 204             'duration': duration,
 205             'view_count': view_count,
 206             'age_limit': age_limit,
 207             'formats': formats,
 208         }