youtube_dl/extractor/nba.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6 )
   7
   8
   9 class NBAIE(InfoExtractor):
  10     _VALID_URL = r'^(?:https?://)?(?:watch\.|www\.)?nba\.com/(?:nba/)?video(/[^?]*?)(?:/index\.html)?(?:\?.*)?$'
  11     _TEST = {
  12         u'url': u'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
  13         u'file': u'0021200253-okc-bkn-recap.nba.mp4',
  14         u'md5': u'c0edcfc37607344e2ff8f13c378c88a4',
  15         u'info_dict': {
  16             u"description": u"Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.",
  17             u"title": u"Thunder vs. Nets"
  18         }
  19     }
  20
  21     def _real_extract(self, url):
  22         mobj = re.match(self._VALID_URL, url)
  23         if mobj is None:
  24             raise ExtractorError(u'Invalid URL: %s' % url)
  25
  26         video_id = mobj.group(1)
  27
  28         webpage = self._download_webpage(url, video_id)
  29
  30         video_url = u'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4'
  31
  32         shortened_video_id = video_id.rpartition('/')[2]
  33         title = self._og_search_title(webpage, default=shortened_video_id).replace('NBA.com: ', '')
  34
  35         # It isn't there in the HTML it returns to us
  36         # uploader_date = self._html_search_regex(r'<b>Date:</b> (.*?)</div>', webpage, 'upload_date', fatal=False)
  37
  38         description = self._html_search_regex(r'<meta name="description" (?:content|value)="(.*?)" />', webpage, 'description', fatal=False)
  39
  40         info = {
  41             'id': shortened_video_id,
  42             'url': video_url,
  43             'ext': 'mp4',
  44             'title': title,
  45             # 'uploader_date': uploader_date,
  46             'description': description,
  47         }
  48         return [info]