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
   8 class CNNIE(InfoExtractor):
   9     _VALID_URL = r'''(?x)https?://((edition|www)\.)?cnn\.com/video/(data/.+?|\?)/
  10         (?P<path>.+?/(?P<title>[^/]+?)(?:\.cnn|(?=&)))'''
  11
  12     _TESTS = [{
  13         u'url': u'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
  14         u'file': u'sports_2013_06_09_nadal-1-on-1.cnn.mp4',
  15         u'md5': u'3e6121ea48df7e2259fe73a0628605c4',
  16         u'info_dict': {
  17             u'title': u'Nadal wins 8th French Open title',
  18             u'description': u'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
  19         },
  20     },
  21     {
  22         u"url": u"http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29",
  23         u"file": u"us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
  24         u"md5": u"b5cc60c60a3477d185af8f19a2a26f4e",
  25         u"info_dict": {
  26             u"title": "Student's epic speech stuns new freshmen",
  27             u"description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\""
  28         }
  29     }]
  30
  31     def _real_extract(self, url):
  32         mobj = re.match(self._VALID_URL, url)
  33         path = mobj.group('path')
  34         page_title = mobj.group('title')
  35         info_url = u'http://cnn.com/video/data/3.0/%s/index.xml' % path
  36         info_xml = self._download_webpage(info_url, page_title)
  37         info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
  38
  39         formats = []
  40         for f in info.findall('files/file'):
  41             mf = re.match(r'(\d+)x(\d+)(?:_(.*)k)?',f.attrib['bitrate'])
  42             if mf is not None:
  43                 formats.append((int(mf.group(1)), int(mf.group(2)), int(mf.group(3) or 0), f.text))
  44         formats = sorted(formats)
  45         (_,_,_, video_path) = formats[-1]
  46         video_url = 'http://ht.cdn.turner.com/cnn/big%s' % video_path
  47
  48         thumbnails = sorted([((int(t.attrib['height']),int(t.attrib['width'])), t.text) for t in info.findall('images/image')])
  49         thumbs_dict = [{'resolution': res, 'url': t_url} for (res, t_url) in thumbnails]
  50
  51         return {'id': info.attrib['id'],
  52                 'title': info.find('headline').text,
  53                 'url': video_url,
  54                 'ext': determine_ext(video_url),
  55                 'thumbnail': thumbnails[-1][1],
  56                 'thumbnails': thumbs_dict,
  57                 'description': info.find('description').text,
  58                 }