youtube_dl/extractor/cnn.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     int_or_none,
   8     parse_duration,
   9 )
  10
  11
  12 class CNNIE(InfoExtractor):
  13     _VALID_URL = r'''(?x)https?://((edition|www)\.)?cnn\.com/video/(data/.+?|\?)/
  14         (?P<path>.+?/(?P<title>[^/]+?)(?:\.cnn|(?=&)))'''
  15
  16     _TESTS = [{
  17         'url': 'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
  18         'file': 'sports_2013_06_09_nadal-1-on-1.cnn.mp4',
  19         'md5': '3e6121ea48df7e2259fe73a0628605c4',
  20         'info_dict': {
  21             'title': 'Nadal wins 8th French Open title',
  22             'description': 'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
  23             'duration': 135,
  24             'upload_date': '20130609',
  25         },
  26     },
  27     {
  28         "url": "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",
  29         "file": "us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
  30         "md5": "b5cc60c60a3477d185af8f19a2a26f4e",
  31         "info_dict": {
  32             "title": "Student's epic speech stuns new freshmen",
  33             "description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"",
  34             "upload_date": "20130821",
  35         }
  36     }]
  37
  38     def _real_extract(self, url):
  39         mobj = re.match(self._VALID_URL, url)
  40         path = mobj.group('path')
  41         page_title = mobj.group('title')
  42         info_url = 'http://cnn.com/video/data/3.0/%s/index.xml' % path
  43         info = self._download_xml(info_url, page_title)
  44
  45         formats = []
  46         rex = re.compile(r'''(?x)
  47             (?P<width>[0-9]+)x(?P<height>[0-9]+)
  48             (?:_(?P<bitrate>[0-9]+)k)?
  49         ''')
  50         for f in info.findall('files/file'):
  51             video_url = 'http://ht.cdn.turner.com/cnn/big%s' % (f.text.strip())
  52             fdct = {
  53                 'format_id': f.attrib['bitrate'],
  54                 'url': video_url,
  55             }
  56
  57             mf = rex.match(f.attrib['bitrate'])
  58             if mf:
  59                 fdct['width'] = int(mf.group('width'))
  60                 fdct['height'] = int(mf.group('height'))
  61                 fdct['tbr'] = int_or_none(mf.group('bitrate'))
  62             else:
  63                 mf = rex.search(f.text)
  64                 if mf:
  65                     fdct['width'] = int(mf.group('width'))
  66                     fdct['height'] = int(mf.group('height'))
  67                     fdct['tbr'] = int_or_none(mf.group('bitrate'))
  68                 else:
  69                     mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
  70                     if mi:
  71                         if mi.group(1) == 'audio':
  72                             fdct['vcodec'] = 'none'
  73                             fdct['ext'] = 'm4a'
  74                         else:
  75                             fdct['tbr'] = int(mi.group(1))
  76
  77             formats.append(fdct)
  78
  79         self._sort_formats(formats)
  80
  81         thumbnails = sorted([((int(t.attrib['height']),int(t.attrib['width'])), t.text) for t in info.findall('images/image')])
  82         thumbs_dict = [{'resolution': res, 'url': t_url} for (res, t_url) in thumbnails]
  83
  84         metas_el = info.find('metas')
  85         upload_date = (
  86             metas_el.attrib.get('version') if metas_el is not None else None)
  87
  88         duration_el = info.find('length')
  89         duration = parse_duration(duration_el.text)
  90
  91         return {
  92             'id': info.attrib['id'],
  93             'title': info.find('headline').text,
  94             'formats': formats,
  95             'thumbnail': thumbnails[-1][1],
  96             'thumbnails': thumbs_dict,
  97             'description': info.find('description').text,
  98             'duration': duration,
  99             'upload_date': upload_date,
 100         }