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         # Regular iframe embedding
  56         mobj = re.search(
  57             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1',
  58             webpage)
  59         if mobj is not None:
  60             return mobj.group('url')
  61         # Basic usage embedding (see http://dultonmedia.github.io/eplayer/)
  62         mobj = re.search(
  63             r'''(?xs)
  64                     <script[^>]+
  65                         src=(?P<q1>["\'])(?:https?:)?//(?P<host>.+?\.media\.eagleplatform\.com)/player/player\.js(?P=q1)
  66                     .+?
  67                     <div[^>]+
  68                         class=(?P<q2>["\'])eagleplayer(?P=q2)[^>]+
  69                         data-id=["\'](?P<id>\d+)
  70             ''', webpage)
  71         if mobj is not None:
  72             return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
  73
  74     @staticmethod
  75     def _handle_error(response):
  76         status = int_or_none(response.get('status', 200))
  77         if status != 200:
  78             raise ExtractorError(' '.join(response['errors']), expected=True)
  79
  80     def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
  81         try:
  82             response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
  83         except ExtractorError as ee:
  84             if isinstance(ee.cause, compat_HTTPError):
  85                 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
  86                 self._handle_error(response)
  87             raise
  88         return response
  89
  90     def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'):
  91         return self._download_json(url_or_request, video_id, note)['data'][0]
  92
  93     def _real_extract(self, url):
  94         mobj = re.match(self._VALID_URL, url)
  95         host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
  96
  97         player_data = self._download_json(
  98             'http://%s/api/player_data?id=%s' % (host, video_id), video_id)
  99
 100         media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
 101
 102         title = media['title']
 103         description = media.get('description')
 104         thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:')
 105         duration = int_or_none(media.get('duration'))
 106         view_count = int_or_none(media.get('views'))
 107
 108         age_restriction = media.get('age_restriction')
 109         age_limit = None
 110         if age_restriction:
 111             age_limit = 0 if age_restriction == 'allow_all' else 18
 112
 113         secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:')
 114
 115         formats = []
 116
 117         m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
 118         m3u8_formats = self._extract_m3u8_formats(
 119             m3u8_url, video_id,
 120             'mp4', entry_protocol='m3u8_native', m3u8_id='hls')
 121         formats.extend(m3u8_formats)
 122
 123         mp4_url = self._get_video_url(
 124             # Secure mp4 URL is constructed according to Player.prototype.mp4 from
 125             # http://lentaru.media.eagleplatform.com/player/player.js
 126             re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4', secure_m3u8),
 127             video_id, 'Downloading mp4 JSON')
 128         mp4_url_basename = url_basename(mp4_url)
 129         for m3u8_format in m3u8_formats:
 130             mobj = re.search('/([^/]+)/index\.m3u8', m3u8_format['url'])
 131             if mobj:
 132                 http_format = m3u8_format.copy()
 133                 video_url = mp4_url.replace(mp4_url_basename, mobj.group(1))
 134                 if not self._is_valid_url(video_url, video_id):
 135                     continue
 136                 http_format.update({
 137                     'url': video_url,
 138                     'format_id': m3u8_format['format_id'].replace('hls', 'http'),
 139                     'protocol': 'http',
 140                 })
 141                 formats.append(http_format)
 142
 143         self._sort_formats(formats)
 144
 145         return {
 146             'id': video_id,
 147             'title': title,
 148             'description': description,
 149             'thumbnail': thumbnail,
 150             'duration': duration,
 151             'view_count': view_count,
 152             'age_limit': age_limit,
 153             'formats': formats,
 154         }