_ Git - youtube-dl/blob - youtube_dl/extractor/mlb.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6
   7
   8 class MlbIE(InfoExtractor):
   9     _VALID_URL = r'http?://m\.mlb\.com/video/topic/[0-9]+/v(?P<id>n?\d+)/.*$'
  10     _TEST = {
  11         'url': 'http://m.mlb.com/video/topic/81536970/v34496663/mianym-stanton-practices-for-the-home-run-derby',
  12         'md5': u'd9c022c10d21f849f49c05ae12a8a7e9',
  13         'info_dict': {
  14             'id': '34496663',
  15             'ext': 'mp4',
  16             'format': 'mp4',
  17             'description': "7/11/14: Giancarlo Stanton practices for the Home Run Derby prior to the game against the Mets",
  18             'title': "Stanton prepares for Derby",
  19         },
  20     }
  21
  22     def _real_extract(self, url):
  23         mobj = re.match(self._VALID_URL, url)
  24         video_id = mobj.group('id')
  25
  26         webpage = self._download_webpage(url, video_id)
  27
  28         title = self._og_search_title(webpage, default=video_id)
  29         description = self._html_search_regex(r'<meta name="description" (?:content|value)="(.*?)"/>', webpage, 'description', fatal=False)
  30         thumbnail = self._html_search_regex(r'<meta itemprop="image" (?:content|value)="(.*?)" />', webpage, 'image', fatal=False)
  31
  32         # use the video_id to find the Media detail XML
  33         id_len = len(video_id)
  34         _mediadetail_url = 'http://m.mlb.com/gen/multimedia/detail/'+video_id[id_len-3]+'/'+video_id[id_len-2]+'/'+video_id[id_len-1]+'/'+video_id+'.xml'
  35
  36         mediadetails = self._download_xml(_mediadetail_url, video_id, "Downloading media detail...")
  37         has1500K = 0
  38         has1200K = 0
  39         has600K = 0
  40         # loop through the list of url's and only get the highest quality MP4 content
  41         for element in mediadetails.findall('url'):
  42             scenario = element.attrib['playback_scenario']
  43             if scenario.startswith(u'FLASH'):
  44                 if scenario.startswith(u'FLASH_1800K'):
  45                     video_url = element.text
  46                     # 1800K is the current highest quality video on MLB.com
  47                     break
  48                 else:
  49                     if scenario.startswith(u'FLASH_1500K'):
  50                         video_url = element.text
  51                         has1500K = 1
  52                     else:
  53                         if (scenario.startswith(u'FLASH_1200K') and not has1500K):
  54                             video_url = element.text
  55                             has1200K = 1
  56                         else:
  57                             if (scenario.startswith(u'FLASH_600K') and not has1200K):
  58                                 video_url = element.text
  59                                 has600K = 1
  60                             else:
  61                                 if (scenario.startswith(u'FLASH_300K') and not has600K):
  62                                     video_url = element.text
  63
  64         return {
  65             'id': video_id,
  66             'url': video_url,
  67             'extractor': 'mlb',
  68             'webpage_url': url,
  69             'title': title,
  70             'ext': 'mp4',
  71             'format': 'mp4',
  72             'description': description,
  73             'thumbnail': thumbnail,
  74         }