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 )
  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': 're:^https?://.*\.jpg$',
  35             'duration': 87,
  36             'view_count': int,
  37             'age_limit': 0,
  38         },
  39     }, {
  40         # http://muz-tv.ru/play/7129/
  41         # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true
  42         'url': 'eagleplatform:media.clipyou.ru:12820',
  43         'md5': '358597369cf8ba56675c1df15e7af624',
  44         'info_dict': {
  45             'id': '12820',
  46             'ext': 'mp4',
  47             'title': "'O Sole Mio",
  48             'thumbnail': 're:^https?://.*\.jpg$',
  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         # Basic usage embedding (see http://dultonmedia.github.io/eplayer/)
  64         mobj = re.search(
  65             r'''(?xs)
  66                     <script[^>]+
  67                         src=(?P<q1>["\'])(?:https?:)?//(?P<host>.+?\.media\.eagleplatform\.com)/player/player\.js(?P=q1)
  68                     .+?
  69                     <div[^>]+
  70                         class=(?P<q2>["\'])eagleplayer(?P=q2)[^>]+
  71                         data-id=["\'](?P<id>\d+)
  72             ''', webpage)
  73         if mobj is not None:
  74             return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
  75
  76     @staticmethod
  77     def _handle_error(response):
  78         status = int_or_none(response.get('status', 200))
  79         if status != 200:
  80             raise ExtractorError(' '.join(response['errors']), expected=True)
  81
  82     def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', *args, **kwargs):
  83         try:
  84             response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
  85         except ExtractorError as ee:
  86             if isinstance(ee.cause, compat_HTTPError):
  87                 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
  88                 self._handle_error(response)
  89             raise
  90         return response
  91
  92     def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'):
  93         return self._download_json(url_or_request, video_id, note)['data'][0]
  94
  95     def _real_extract(self, url):
  96         mobj = re.match(self._VALID_URL, url)
  97         host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
  98
  99         player_data = self._download_json(
 100             'http://%s/api/player_data?id=%s' % (host, video_id), video_id)
 101
 102         media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
 103
 104         title = media['title']
 105         description = media.get('description')
 106         thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:')
 107         duration = int_or_none(media.get('duration'))
 108         view_count = int_or_none(media.get('views'))
 109
 110         age_restriction = media.get('age_restriction')
 111         age_limit = None
 112         if age_restriction:
 113             age_limit = 0 if age_restriction == 'allow_all' else 18
 114
 115         secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:')
 116
 117         formats = []
 118
 119         m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
 120         m3u8_formats = self._extract_m3u8_formats(
 121             m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
 122             m3u8_id='hls', fatal=False)
 123         formats.extend(m3u8_formats)
 124
 125         m3u8_formats_dict = {}
 126         for f in m3u8_formats:
 127             if f.get('height') is not None:
 128                 m3u8_formats_dict[f['height']] = f
 129
 130         mp4_data = self._download_json(
 131             # Secure mp4 URL is constructed according to Player.prototype.mp4 from
 132             # http://lentaru.media.eagleplatform.com/player/player.js
 133             re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4s', secure_m3u8),
 134             video_id, 'Downloading mp4 JSON', fatal=False)
 135         if mp4_data:
 136             for format_id, format_url in mp4_data.get('data', {}).items():
 137                 if not isinstance(format_url, compat_str):
 138                     continue
 139                 height = int_or_none(format_id)
 140                 if height is not None and m3u8_formats_dict.get(height):
 141                     f = m3u8_formats_dict[height].copy()
 142                     f.update({
 143                         'format_id': f['format_id'].replace('hls', 'http'),
 144                         'protocol': 'http',
 145                     })
 146                 else:
 147                     f = {
 148                         'format_id': 'http-%s' % format_id,
 149                         'height': int_or_none(format_id),
 150                     }
 151                 f['url'] = format_url
 152                 formats.append(f)
 153
 154         self._sort_formats(formats)
 155
 156         return {
 157             'id': video_id,
 158             'title': title,
 159             'description': description,
 160             'thumbnail': thumbnail,
 161             'duration': duration,
 162             'view_count': view_count,
 163             'age_limit': age_limit,
 164             'formats': formats,
 165         }