_ Git - youtube-dl/blob - 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|(?=&)))'''
  16
  17     _TESTS = [{
  18         'url': 'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
  19         'file': 'sports_2013_06_09_nadal-1-on-1.cnn.mp4',
  20         'md5': '3e6121ea48df7e2259fe73a0628605c4',
  21         'info_dict': {
  22             'title': 'Nadal wins 8th French Open title',
  23             'description': 'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
  24             'duration': 135,
  25             'upload_date': '20130609',
  26         },
  27     },
  28     {
  29         "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",
  30         "file": "us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
  31         "md5": "b5cc60c60a3477d185af8f19a2a26f4e",
  32         "info_dict": {
  33             "title": "Student's epic speech stuns new freshmen",
  34             "description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"",
  35             "upload_date": "20130821",
  36         }
  37     }]
  38
  39     def _real_extract(self, url):
  40         mobj = re.match(self._VALID_URL, url)
  41         path = mobj.group('path')
  42         page_title = mobj.group('title')
  43         info_url = 'http://cnn.com/video/data/3.0/%s/index.xml' % path
  44         info = self._download_xml(info_url, page_title)
  45
  46         formats = []
  47         rex = re.compile(r'''(?x)
  48             (?P<width>[0-9]+)x(?P<height>[0-9]+)
  49             (?:_(?P<bitrate>[0-9]+)k)?
  50         ''')
  51         for f in info.findall('files/file'):
  52             video_url = 'http://ht.cdn.turner.com/cnn/big%s' % (f.text.strip())
  53             fdct = {
  54                 'format_id': f.attrib['bitrate'],
  55                 'url': video_url,
  56             }
  57
  58             mf = rex.match(f.attrib['bitrate'])
  59             if mf:
  60                 fdct['width'] = int(mf.group('width'))
  61                 fdct['height'] = int(mf.group('height'))
  62                 fdct['tbr'] = int_or_none(mf.group('bitrate'))
  63             else:
  64                 mf = rex.search(f.text)
  65                 if mf:
  66                     fdct['width'] = int(mf.group('width'))
  67                     fdct['height'] = int(mf.group('height'))
  68                     fdct['tbr'] = int_or_none(mf.group('bitrate'))
  69                 else:
  70                     mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
  71                     if mi:
  72                         if mi.group(1) == 'audio':
  73                             fdct['vcodec'] = 'none'
  74                             fdct['ext'] = 'm4a'
  75                         else:
  76                             fdct['tbr'] = int(mi.group(1))
  77
  78             formats.append(fdct)
  79
  80         self._sort_formats(formats)
  81
  82         thumbnails = sorted([((int(t.attrib['height']),int(t.attrib['width'])), t.text) for t in info.findall('images/image')])
  83         thumbs_dict = [{'resolution': res, 'url': t_url} for (res, t_url) in thumbnails]
  84
  85         metas_el = info.find('metas')
  86         upload_date = (
  87             metas_el.attrib.get('version') if metas_el is not None else None)
  88
  89         duration_el = info.find('length')
  90         duration = parse_duration(duration_el.text)
  91
  92         return {
  93             'id': info.attrib['id'],
  94             'title': info.find('headline').text,
  95             'formats': formats,
  96             'thumbnail': thumbnails[-1][1],
  97             'thumbnails': thumbs_dict,
  98             'description': info.find('description').text,
  99             'duration': duration,
 100             'upload_date': upload_date,
 101         }
 102
 103
 104 class CNNBlogsIE(InfoExtractor):
 105     _VALID_URL = r'https?://[^\.]+\.blogs\.cnn\.com/.+'
 106     _TEST = {
 107         'url': 'http://reliablesources.blogs.cnn.com/2014/02/09/criminalizing-journalism/',
 108         'md5': '3e56f97b0b6ffb4b79f4ea0749551084',
 109         'info_dict': {
 110             'id': 'bestoftv/2014/02/09/criminalizing-journalism.cnn',
 111             'ext': 'mp4',
 112             'title': 'Criminalizing journalism?',
 113             'description': 'Glenn Greenwald responds to comments made this week on Capitol Hill that journalists could be criminal accessories.',
 114             'upload_date': '20140209',
 115         },
 116         'add_ie': ['CNN'],
 117     }
 118
 119     def _real_extract(self, url):
 120         webpage = self._download_webpage(url, url_basename(url))
 121         cnn_url = self._html_search_regex(r'data-url="(.+?)"', webpage, 'cnn url')
 122         return {
 123             '_type': 'url',
 124             'url': cnn_url,
 125             'ie_key': CNNIE.ie_key(),
 126         }