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     url_basename,
  10 )
  11
  12
  13 class CNNIE(InfoExtractor):
  14     _VALID_URL = r'''(?x)https?://((edition|www)\.)?cnn\.com/video/(data/.+?|\?)/
  15         (?P<path>.+?/(?P<title>[^/]+?)(?:\.cnn(-ap)?|(?=&)))'''
  16
  17     _TESTS = [{
  18         'url': 'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
  19         'md5': '3e6121ea48df7e2259fe73a0628605c4',
  20         'info_dict': {
  21             'id': 'sports_2013_06_09_nadal-1-on-1.cnn',
  22             'ext': 'mp4',
  23             'title': 'Nadal wins 8th French Open title',
  24             'description': 'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
  25             'duration': 135,
  26             'upload_date': '20130609',
  27         },
  28     },
  29     {
  30         "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",
  31         "md5": "b5cc60c60a3477d185af8f19a2a26f4e",
  32         "info_dict": {
  33             'id': 'us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology',
  34             'ext': 'mp4',
  35             "title": "Student's epic speech stuns new freshmen",
  36             "description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"",
  37             "upload_date": "20130821",
  38         }
  39     }]
  40
  41     def _real_extract(self, url):
  42         mobj = re.match(self._VALID_URL, url)
  43         path = mobj.group('path')
  44         page_title = mobj.group('title')
  45         info_url = 'http://cnn.com/video/data/3.0/%s/index.xml' % path
  46         info = self._download_xml(info_url, page_title)
  47
  48         formats = []
  49         rex = re.compile(r'''(?x)
  50             (?P<width>[0-9]+)x(?P<height>[0-9]+)
  51             (?:_(?P<bitrate>[0-9]+)k)?
  52         ''')
  53         for f in info.findall('files/file'):
  54             video_url = 'http://ht.cdn.turner.com/cnn/big%s' % (f.text.strip())
  55             fdct = {
  56                 'format_id': f.attrib['bitrate'],
  57                 'url': video_url,
  58             }
  59
  60             mf = rex.match(f.attrib['bitrate'])
  61             if mf:
  62                 fdct['width'] = int(mf.group('width'))
  63                 fdct['height'] = int(mf.group('height'))
  64                 fdct['tbr'] = int_or_none(mf.group('bitrate'))
  65             else:
  66                 mf = rex.search(f.text)
  67                 if mf:
  68                     fdct['width'] = int(mf.group('width'))
  69                     fdct['height'] = int(mf.group('height'))
  70                     fdct['tbr'] = int_or_none(mf.group('bitrate'))
  71                 else:
  72                     mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
  73                     if mi:
  74                         if mi.group(1) == 'audio':
  75                             fdct['vcodec'] = 'none'
  76                             fdct['ext'] = 'm4a'
  77                         else:
  78                             fdct['tbr'] = int(mi.group(1))
  79
  80             formats.append(fdct)
  81
  82         self._sort_formats(formats)
  83
  84         thumbnails = [{
  85             'height': int(t.attrib['height']),
  86             'width': int(t.attrib['width']),
  87             'url': t.text,
  88         } for t in info.findall('images/image')]
  89
  90         metas_el = info.find('metas')
  91         upload_date = (
  92             metas_el.attrib.get('version') if metas_el is not None else None)
  93
  94         duration_el = info.find('length')
  95         duration = parse_duration(duration_el.text)
  96
  97         return {
  98             'id': info.attrib['id'],
  99             'title': info.find('headline').text,
 100             'formats': formats,
 101             'thumbnails': thumbnails,
 102             'description': info.find('description').text,
 103             'duration': duration,
 104             'upload_date': upload_date,
 105         }
 106
 107
 108 class CNNBlogsIE(InfoExtractor):
 109     _VALID_URL = r'https?://[^\.]+\.blogs\.cnn\.com/.+'
 110     _TEST = {
 111         'url': 'http://reliablesources.blogs.cnn.com/2014/02/09/criminalizing-journalism/',
 112         'md5': '3e56f97b0b6ffb4b79f4ea0749551084',
 113         'info_dict': {
 114             'id': 'bestoftv/2014/02/09/criminalizing-journalism.cnn',
 115             'ext': 'mp4',
 116             'title': 'Criminalizing journalism?',
 117             'description': 'Glenn Greenwald responds to comments made this week on Capitol Hill that journalists could be criminal accessories.',
 118             'upload_date': '20140209',
 119         },
 120         'add_ie': ['CNN'],
 121     }
 122
 123     def _real_extract(self, url):
 124         webpage = self._download_webpage(url, url_basename(url))
 125         cnn_url = self._html_search_regex(r'data-url="(.+?)"', webpage, 'cnn url')
 126         return {
 127             '_type': 'url',
 128             'url': cnn_url,
 129             'ie_key': CNNIE.ie_key(),
 130         }