Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/mailru.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     remove_end,
  10 )
  11
  12
  13 class MailRuIE(InfoExtractor):
  14     IE_NAME = 'mailru'
  15     IE_DESC = 'Видео@Mail.Ru'
  16     _VALID_URL = r'''(?x)
  17                     https?://
  18                         (?:(?:www|m)\.)?my\.mail\.ru/
  19                         (?:
  20                             video/.*\#video=/?(?P<idv1>(?:[^/]+/){3}\d+)|
  21                             (?:(?P<idv2prefix>(?:[^/]+/){2})video/(?P<idv2suffix>[^/]+/\d+))\.html|
  22                             (?:video/embed|\+/video/meta)/(?P<metaid>\d+)
  23                         )
  24                     '''
  25     _TESTS = [
  26         {
  27             'url': 'http://my.mail.ru/video/top#video=/mail/sonypicturesrus/75/76',
  28             'md5': 'dea205f03120046894db4ebb6159879a',
  29             'info_dict': {
  30                 'id': '46301138_76',
  31                 'ext': 'mp4',
  32                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
  33                 'timestamp': 1393235077,
  34                 'upload_date': '20140224',
  35                 'uploader': 'sonypicturesrus',
  36                 'uploader_id': 'sonypicturesrus@mail.ru',
  37                 'duration': 184,
  38             },
  39             'skip': 'Not accessible from Travis CI server',
  40         },
  41         {
  42             'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html',
  43             'md5': '00a91a58c3402204dcced523777b475f',
  44             'info_dict': {
  45                 'id': '46843144_1263',
  46                 'ext': 'mp4',
  47                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
  48                 'timestamp': 1397039888,
  49                 'upload_date': '20140409',
  50                 'uploader': 'hitech',
  51                 'uploader_id': 'hitech@corp.mail.ru',
  52                 'duration': 245,
  53             },
  54             'skip': 'Not accessible from Travis CI server',
  55         },
  56         {
  57             # only available via metaUrl API
  58             'url': 'http://my.mail.ru/mail/720pizle/video/_myvideo/502.html',
  59             'md5': '3b26d2491c6949d031a32b96bd97c096',
  60             'info_dict': {
  61                 'id': '56664382_502',
  62                 'ext': 'mp4',
  63                 'title': ':8336',
  64                 'timestamp': 1449094163,
  65                 'upload_date': '20151202',
  66                 'uploader': '720pizle@mail.ru',
  67                 'uploader_id': '720pizle@mail.ru',
  68                 'duration': 6001,
  69             },
  70             'skip': 'Not accessible from Travis CI server',
  71         },
  72         {
  73             'url': 'http://m.my.mail.ru/mail/3sktvtr/video/_myvideo/138.html',
  74             'only_matching': True,
  75         },
  76         {
  77             'url': 'https://my.mail.ru/video/embed/7949340477499637815',
  78             'only_matching': True,
  79         },
  80         {
  81             'url': 'http://my.mail.ru/+/video/meta/7949340477499637815',
  82             'only_matching': True,
  83         }
  84     ]
  85
  86     def _real_extract(self, url):
  87         mobj = re.match(self._VALID_URL, url)
  88         meta_id = mobj.group('metaid')
  89
  90         video_id = None
  91         if meta_id:
  92             meta_url = 'https://my.mail.ru/+/video/meta/%s' % meta_id
  93         else:
  94             video_id = mobj.group('idv1')
  95             if not video_id:
  96                 video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
  97             webpage = self._download_webpage(url, video_id)
  98             page_config = self._parse_json(self._search_regex(
  99                 r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>',
 100                 webpage, 'page config', default='{}'), video_id, fatal=False)
 101             if page_config:
 102                 meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl')
 103             else:
 104                 meta_url = None
 105
 106         video_data = None
 107         if meta_url:
 108             video_data = self._download_json(
 109                 meta_url, video_id or meta_id, 'Downloading video meta JSON',
 110                 fatal=not video_id)
 111
 112         # Fallback old approach
 113         if not video_data:
 114             video_data = self._download_json(
 115                 'http://api.video.mail.ru/videos/%s.json?new=1' % video_id,
 116                 video_id, 'Downloading video JSON')
 117
 118         formats = []
 119         for f in video_data['videos']:
 120             video_url = f.get('url')
 121             if not video_url:
 122                 continue
 123             format_id = f.get('key')
 124             height = int_or_none(self._search_regex(
 125                 r'^(\d+)[pP]$', format_id, 'height', default=None)) if format_id else None
 126             formats.append({
 127                 'url': video_url,
 128                 'format_id': format_id,
 129                 'height': height,
 130             })
 131         self._sort_formats(formats)
 132
 133         meta_data = video_data['meta']
 134         title = remove_end(meta_data['title'], '.mp4')
 135
 136         author = video_data.get('author')
 137         uploader = author.get('name')
 138         uploader_id = author.get('id') or author.get('email')
 139         view_count = int_or_none(video_data.get('viewsCount') or video_data.get('views_count'))
 140
 141         acc_id = meta_data.get('accId')
 142         item_id = meta_data.get('itemId')
 143         content_id = '%s_%s' % (acc_id, item_id) if acc_id and item_id else video_id
 144
 145         thumbnail = meta_data.get('poster')
 146         duration = int_or_none(meta_data.get('duration'))
 147         timestamp = int_or_none(meta_data.get('timestamp'))
 148
 149         return {
 150             'id': content_id,
 151             'title': title,
 152             'thumbnail': thumbnail,
 153             'timestamp': timestamp,
 154             'uploader': uploader,
 155             'uploader_id': uploader_id,
 156             'duration': duration,
 157             'view_count': view_count,
 158             'formats': formats,
 159         }