youtube_dl/extractor/mailru.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8
   9 class MailRuIE(InfoExtractor):
  10     IE_NAME = 'mailru'
  11     IE_DESC = 'Видео@Mail.Ru'
  12     _VALID_URL = r'http://(?:www\.)?my\.mail\.ru/(?:video/.*#video=/?(?P<idv1>(?:[^/]+/){3}\d+)|(?:(?P<idv2prefix>(?:[^/]+/){2})video/(?P<idv2suffix>[^/]+/\d+))\.html)'
  13
  14     _TESTS = [
  15         {
  16             'url': 'http://my.mail.ru/video/top#video=/mail/sonypicturesrus/75/76',
  17             'md5': 'dea205f03120046894db4ebb6159879a',
  18             'info_dict': {
  19                 'id': '46301138_76',
  20                 'ext': 'mp4',
  21                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
  22                 'timestamp': 1393232740,
  23                 'upload_date': '20140224',
  24                 'uploader': 'sonypicturesrus',
  25                 'uploader_id': 'sonypicturesrus@mail.ru',
  26                 'duration': 184,
  27             },
  28             'skip': 'Not accessible from Travis CI server',
  29         },
  30         {
  31             'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html',
  32             'md5': '00a91a58c3402204dcced523777b475f',
  33             'info_dict': {
  34                 'id': '46843144_1263',
  35                 'ext': 'mp4',
  36                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
  37                 'timestamp': 1397217632,
  38                 'upload_date': '20140411',
  39                 'uploader': 'hitech',
  40                 'uploader_id': 'hitech@corp.mail.ru',
  41                 'duration': 245,
  42             },
  43             'skip': 'Not accessible from Travis CI server',
  44         },
  45         {
  46             # only available via metaUrl API
  47             'url': 'http://my.mail.ru/mail/720pizle/video/_myvideo/502.html',
  48             'md5': '3b26d2491c6949d031a32b96bd97c096',
  49             'info_dict': {
  50                 'id': '56664382_502',
  51                 'ext': 'mp4',
  52                 'title': ':8336',
  53                 'timestamp': 1449094163,
  54                 'upload_date': '20151202',
  55                 'uploader': '720pizle@mail.ru',
  56                 'uploader_id': '720pizle@mail.ru',
  57                 'duration': 6001,
  58             },
  59             'skip': 'Not accessible from Travis CI server',
  60         }
  61     ]
  62
  63     def _real_extract(self, url):
  64         mobj = re.match(self._VALID_URL, url)
  65         video_id = mobj.group('idv1')
  66
  67         if not video_id:
  68             video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
  69
  70         webpage = self._download_webpage(url, video_id)
  71
  72         video_data = None
  73
  74         page_config = self._parse_json(self._search_regex(
  75             r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>',
  76             webpage, 'page config', default='{}'), video_id, fatal=False)
  77         if page_config:
  78             meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl')
  79             if meta_url:
  80                 video_data = self._download_json(
  81                     meta_url, video_id, 'Downloading video meta JSON', fatal=False)
  82
  83         # Fallback old approach
  84         if not video_data:
  85             video_data = self._download_json(
  86                 'http://api.video.mail.ru/videos/%s.json?new=1' % video_id,
  87                 video_id, 'Downloading video JSON')
  88
  89         author = video_data['author']
  90         uploader = author['name']
  91         uploader_id = author.get('id') or author.get('email')
  92         view_count = video_data.get('views_count')
  93
  94         meta_data = video_data['meta']
  95         content_id = '%s_%s' % (
  96             meta_data.get('accId', ''), meta_data['itemId'])
  97         title = meta_data['title']
  98         if title.endswith('.mp4'):
  99             title = title[:-4]
 100         thumbnail = meta_data['poster']
 101         duration = meta_data['duration']
 102         timestamp = meta_data['timestamp']
 103
 104         formats = [
 105             {
 106                 'url': video['url'],
 107                 'format_id': video['key'],
 108                 'height': int(video['key'].rstrip('p'))
 109             } for video in video_data['videos']
 110         ]
 111         self._sort_formats(formats)
 112
 113         return {
 114             'id': content_id,
 115             'title': title,
 116             'thumbnail': thumbnail,
 117             'timestamp': timestamp,
 118             'uploader': uploader,
 119             'uploader_id': uploader_id,
 120             'duration': duration,
 121             'view_count': view_count,
 122             'formats': formats,
 123         }