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

   1 import re
   2 import xml.etree.ElementTree
   3
   4 from .common import InfoExtractor
   5 from ..utils import determine_ext
   6
   7 class CNNIE(InfoExtractor):
   8     _VALID_URL = r'https?://(edition\.)?cnn\.com/video/(data/.+?|\?)/(?P<path>.+?/(?P<title>[^/]+?)\.cnn)'
   9
  10     _TEST = {
  11         u'url': u'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
  12         u'file': u'sports_2013_06_09_nadal-1-on-1.cnn.mp4',
  13         u'md5': u'3e6121ea48df7e2259fe73a0628605c4',
  14         u'info_dict': {
  15             u'title': u'Nadal wins 8th French Open title',
  16             u'description': u'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
  17         },
  18     }
  19
  20     def _real_extract(self, url):
  21         mobj = re.match(self._VALID_URL, url)
  22         path = mobj.group('path')
  23         page_title = mobj.group('title')
  24         info_xml = self._download_webpage(
  25             'http://cnn.com/video/data/3.0/%s/index.xml' % path, page_title)
  26         info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
  27
  28         formats = []
  29         for f in info.findall('files/file'):
  30             mf = re.match(r'(\d+)x(\d+)(?:_(.*)k)?',f.attrib['bitrate'])
  31             if mf is not None:
  32                 formats.append((int(mf.group(1)), int(mf.group(2)), int(mf.group(3) or 0), f.text))
  33         formats = sorted(formats)
  34         (_,_,_, video_path) = formats[-1]
  35         video_url = 'http://ht.cdn.turner.com/cnn/big%s' % video_path
  36
  37         thumbnails = sorted([((int(t.attrib['height']),int(t.attrib['width'])), t.text) for t in info.findall('images/image')])
  38         thumbs_dict = [{'resolution': res, 'url': t_url} for (res, t_url) in thumbnails]
  39
  40         return {'id': info.attrib['id'],
  41                 'title': info.find('headline').text,
  42                 'url': video_url,
  43                 'ext': determine_ext(video_url),
  44                 'thumbnail': thumbnails[-1][1],
  45                 'thumbnails': thumbs_dict,
  46                 'description': info.find('description').text,
  47                 }