_ 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+)/.*$'
  10     _TEST = {
  11         'url': 'http://m.mlb.com/video/topic/81536970/v34496663/mianym-stanton-practices-for-the-home-run-derby',
  12         'md5': u'd9c022c10d21f849f49c05ae12a8a7e9',
  13         'info_dict': {
  14             'id': '34496663',
  15             'ext': 'mp4',
  16             'format': 'mp4',
  17             'description': "7/11/14: Giancarlo Stanton practices for the Home Run Derby prior to the game against the Mets",
  18             'title': "Stanton prepares for Derby",
  19         },
  20     }
  21
  22     def _real_extract(self, url):
  23         mobj = re.match(self._VALID_URL, url)
  24         video_id = mobj.group('id')
  25
  26         webpage = self._download_webpage(url, video_id)
  27
  28         title = self._og_search_title(webpage, default=video_id)
  29         description = self._html_search_regex(r'<meta name="description" (?:content|value)="(.*?)"/>', webpage, 'description', fatal=False)
  30         thumbnail = self._html_search_regex(r'<meta itemprop="image" (?:content|value)="(.*?)" />', webpage, 'image', fatal=False)
  31
  32         # use the thumbnail URL to find the folder that contains the videos
  33         _image_url = r'http://mediadownloads.mlb.com/mlbam/(?P<_date>n?.+)/images/.*$'
  34         bobj = re.match(_image_url, thumbnail)
  35         datestr = bobj.group('_date')
  36         base_url = 'http://mediadownloads.mlb.com/mlbam/' + datestr
  37         filespage = self._download_webpage(base_url, video_id)
  38
  39         # Try 1800K, 1500K, 1200K, 600K, then 300K videos
  40         video = self._html_search_regex(r'<li><a href="(.*?)_'+video_id+'_1800K.mp4"', filespage, '1800K', fatal=False)
  41         if video is not None:
  42             video_url = base_url+'/'+video+'_'+video_id+'_1800K.mp4'
  43         else:
  44             video = self._html_search_regex(r'<li><a href="(.*?)_'+video_id+'_1500K.mp4"', filespage, '1500K', fatal=False)
  45             if video is not None:
  46                 video_url = base_url+'/'+video+'_'+video_id+'_1500K.mp4'
  47             else:
  48                 video = self._html_search_regex(r'<li><a href="(.*?)_'+video_id+'_600K.mp4"', filespage, '600K', fatal=False)
  49                 if video is not None:
  50                     video_url = base_url+'/'+video+'_'+video_id+'_600K.mp4'
  51                 else:
  52                     video = self._html_search_regex(r'<li><a href="(.*?)_'+video_id+'_300K.mp4"', filespage, 'MLB', fatal=False)
  53                     if video is not None:
  54                         video_url = base_url+'/'+video+'_'+video_id+'_300K.mp4'
  55                     else:
  56                         # nothing valuable to return
  57                         return None
  58
  59         return {
  60             'id': video_id,
  61             'url': video_url,
  62             'title': title,
  63             'ext': 'mp4',
  64             'format': 'mp4',
  65             'description': description,
  66             'thumbnail': thumbnail,
  67         }