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 compat_HTTPError
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     url_basename,
  12 )
  13
  14
  15 class EaglePlatformIE(InfoExtractor):
  16     _VALID_URL = r'''(?x)
  17                     (?:
  18                         eagleplatform:(?P<custom_host>[^/]+):|
  19                         https?://(?P<host>.+?\.media\.eagleplatform\.com)/index/player\?.*\brecord_id=
  20                     )
  21                     (?P<id>\d+)
  22                 '''
  23     _TESTS = [{
  24         # http://lenta.ru/news/2015/03/06/navalny/
  25         'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',
  26         # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
  27         'info_dict': {
  28             'id': '227304',
  29             'ext': 'mp4',
  30             'title': 'Навальный вышел на свободу',
  31             'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
  32             'thumbnail': 're:^https?://.*\.jpg$',
  33             'duration': 87,
  34             'view_count': int,
  35             'age_limit': 0,
  36         },
  37     }, {
  38         # http://muz-tv.ru/play/7129/
  39         # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true
  40         'url': 'eagleplatform:media.clipyou.ru:12820',
  41         'md5': '358597369cf8ba56675c1df15e7af624',
  42         'info_dict': {
  43             'id': '12820',
  44             'ext': 'mp4',
  45             'title': "'O Sole Mio",
  46             'thumbnail': 're:^https?://.*\.jpg$',
  47             'duration': 216,
  48             'view_count': int,
  49         },
  50         'skip': 'Georestricted',
  51     }]
  52
  53     @staticmethod
  54     def _extract_url(webpage):
  55         mobj = re.search(
  56             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1',
  57             webpage)
  58         if mobj is not None:
  59             return mobj.group('url')
  60
  61     @staticmethod
  62     def _handle_error(response):
  63         status = int_or_none(response.get('status', 200))
  64         if status != 200:
  65             raise ExtractorError(' '.join(response['errors']), expected=True)
  66
  67     def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
  68         try:
  69             response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
  70         except ExtractorError as ee:
  71             if isinstance(ee.cause, compat_HTTPError):
  72                 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
  73                 self._handle_error(response)
  74             raise
  75         return response
  76
  77     def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'):
  78         return self._download_json(url_or_request, video_id, note)['data'][0]
  79
  80     def _real_extract(self, url):
  81         mobj = re.match(self._VALID_URL, url)
  82         host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
  83
  84         player_data = self._download_json(
  85             'http://%s/api/player_data?id=%s' % (host, video_id), video_id)
  86
  87         media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
  88
  89         title = media['title']
  90         description = media.get('description')
  91         thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:')
  92         duration = int_or_none(media.get('duration'))
  93         view_count = int_or_none(media.get('views'))
  94
  95         age_restriction = media.get('age_restriction')
  96         age_limit = None
  97         if age_restriction:
  98             age_limit = 0 if age_restriction == 'allow_all' else 18
  99
 100         secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:')
 101
 102         formats = []
 103
 104         m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
 105         m3u8_formats = self._extract_m3u8_formats(
 106             m3u8_url, video_id,
 107             'mp4', entry_protocol='m3u8_native', m3u8_id='hls')
 108         formats.extend(m3u8_formats)
 109
 110         mp4_url = self._get_video_url(
 111             # Secure mp4 URL is constructed according to Player.prototype.mp4 from
 112             # http://lentaru.media.eagleplatform.com/player/player.js
 113             re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4', secure_m3u8),
 114             video_id, 'Downloading mp4 JSON')
 115         mp4_url_basename = url_basename(mp4_url)
 116         for m3u8_format in m3u8_formats:
 117             mobj = re.search('/([^/]+)/index\.m3u8', m3u8_format['url'])
 118             if mobj:
 119                 http_format = m3u8_format.copy()
 120                 video_url = mp4_url.replace(mp4_url_basename, mobj.group(1))
 121                 if not self._is_valid_url(video_url, video_id):
 122                     continue
 123                 http_format.update({
 124                     'url': video_url,
 125                     'format_id': m3u8_format['format_id'].replace('hls', 'http'),
 126                     'protocol': 'http',
 127                 })
 128                 formats.append(http_format)
 129
 130         self._sort_formats(formats)
 131
 132         return {
 133             'id': video_id,
 134             'title': title,
 135             'description': description,
 136             'thumbnail': thumbnail,
 137             'duration': duration,
 138             'view_count': view_count,
 139             'age_limit': age_limit,
 140             'formats': formats,
 141         }