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'https?://(?:(?:www|m)\.)?my\.mail\.ru/(?:video/.*#video=/?(?P<idv1>(?:[^/]+/){3}\d+)|(?:(?P<idv2prefix>(?:[^/]+/){2})video/(?P<idv2suffix>[^/]+/\d+))\.html)'
  17
  18     _TESTS = [
  19         {
  20             'url': 'http://my.mail.ru/video/top#video=/mail/sonypicturesrus/75/76',
  21             'md5': 'dea205f03120046894db4ebb6159879a',
  22             'info_dict': {
  23                 'id': '46301138_76',
  24                 'ext': 'mp4',
  25                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
  26                 'timestamp': 1393232740,
  27                 'upload_date': '20140224',
  28                 'uploader': 'sonypicturesrus',
  29                 'uploader_id': 'sonypicturesrus@mail.ru',
  30                 'duration': 184,
  31             },
  32             'skip': 'Not accessible from Travis CI server',
  33         },
  34         {
  35             'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html',
  36             'md5': '00a91a58c3402204dcced523777b475f',
  37             'info_dict': {
  38                 'id': '46843144_1263',
  39                 'ext': 'mp4',
  40                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
  41                 'timestamp': 1397039888,
  42                 'upload_date': '20140409',
  43                 'uploader': 'hitech@corp.mail.ru',
  44                 'uploader_id': 'hitech@corp.mail.ru',
  45                 'duration': 245,
  46             },
  47             'skip': 'Not accessible from Travis CI server',
  48         },
  49         {
  50             # only available via metaUrl API
  51             'url': 'http://my.mail.ru/mail/720pizle/video/_myvideo/502.html',
  52             'md5': '3b26d2491c6949d031a32b96bd97c096',
  53             'info_dict': {
  54                 'id': '56664382_502',
  55                 'ext': 'mp4',
  56                 'title': ':8336',
  57                 'timestamp': 1449094163,
  58                 'upload_date': '20151202',
  59                 'uploader': '720pizle@mail.ru',
  60                 'uploader_id': '720pizle@mail.ru',
  61                 'duration': 6001,
  62             },
  63             'skip': 'Not accessible from Travis CI server',
  64         },
  65         {
  66             'url': 'http://m.my.mail.ru/mail/3sktvtr/video/_myvideo/138.html',
  67             'only_matching': True,
  68         }
  69     ]
  70
  71     def _real_extract(self, url):
  72         mobj = re.match(self._VALID_URL, url)
  73         video_id = mobj.group('idv1')
  74
  75         if not video_id:
  76             video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
  77
  78         webpage = self._download_webpage(url, video_id)
  79
  80         video_data = None
  81
  82         page_config = self._parse_json(self._search_regex(
  83             r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>',
  84             webpage, 'page config', default='{}'), video_id, fatal=False)
  85         if page_config:
  86             meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl')
  87             if meta_url:
  88                 video_data = self._download_json(
  89                     meta_url, video_id, 'Downloading video meta JSON', fatal=False)
  90
  91         # Fallback old approach
  92         if not video_data:
  93             video_data = self._download_json(
  94                 'http://api.video.mail.ru/videos/%s.json?new=1' % video_id,
  95                 video_id, 'Downloading video JSON')
  96
  97         formats = []
  98         for f in video_data['videos']:
  99             video_url = f.get('url')
 100             if not video_url:
 101                 continue
 102             format_id = f.get('key')
 103             height = int_or_none(self._search_regex(
 104                 r'^(\d+)[pP]$', format_id, 'height', default=None)) if format_id else None
 105             formats.append({
 106                 'url': video_url,
 107                 'format_id': format_id,
 108                 'height': height,
 109             })
 110         self._sort_formats(formats)
 111
 112         meta_data = video_data['meta']
 113         title = remove_end(meta_data['title'], '.mp4')
 114
 115         author = video_data.get('author')
 116         uploader = author.get('name')
 117         uploader_id = author.get('id') or author.get('email')
 118         view_count = int_or_none(video_data.get('viewsCount') or video_data.get('views_count'))
 119
 120         acc_id = meta_data.get('accId')
 121         item_id = meta_data.get('itemId')
 122         content_id = '%s_%s' % (acc_id, item_id) if acc_id and item_id else video_id
 123
 124         thumbnail = meta_data.get('poster')
 125         duration = int_or_none(meta_data.get('duration'))
 126         timestamp = int_or_none(meta_data.get('timestamp'))
 127
 128         return {
 129             'id': content_id,
 130             'title': title,
 131             'thumbnail': thumbnail,
 132             'timestamp': timestamp,
 133             'uploader': uploader,
 134             'uploader_id': uploader_id,
 135             'duration': duration,
 136             'view_count': view_count,
 137             'formats': formats,
 138         }