Raphaël G. Git Repositories - youtubedl/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 )
  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         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",
  29         u"file": u"us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
  30         u"md5": u"b5cc60c60a3477d185af8f19a2a26f4e",
  31         u"info_dict": {
  32             u"title": "Student's epic speech stuns new freshmen",
  33             u"description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\""
  34         }
  35     }]
  36
  37     def _real_extract(self, url):
  38         mobj = re.match(self._VALID_URL, url)
  39         path = mobj.group('path')
  40         page_title = mobj.group('title')
  41         info_url = 'http://cnn.com/video/data/3.0/%s/index.xml' % path
  42         info = self._download_xml(info_url, page_title)
  43
  44         formats = []
  45         rex = re.compile(r'''(?x)
  46             (?P<width>[0-9]+)x(?P<height>[0-9]+)
  47             (?:_(?P<bitrate>[0-9]+)k)?
  48         ''')
  49         for f in info.findall('files/file'):
  50             video_url = 'http://ht.cdn.turner.com/cnn/big%s' % (f.text.strip())
  51             fdct = {
  52                 'format_id': f.attrib['bitrate'],
  53                 'url': video_url,
  54             }
  55
  56             mf = rex.match(f.attrib['bitrate'])
  57             if mf:
  58                 fdct['width'] = int(mf.group('width'))
  59                 fdct['height'] = int(mf.group('height'))
  60                 fdct['tbr'] = int_or_none(mf.group('bitrate'))
  61             else:
  62                 mf = rex.search(f.text)
  63                 if mf:
  64                     fdct['width'] = int(mf.group('width'))
  65                     fdct['height'] = int(mf.group('height'))
  66                     fdct['tbr'] = int_or_none(mf.group('bitrate'))
  67                 else:
  68                     mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
  69                     if mi:
  70                         if mi.group(1) == 'audio':
  71                             fdct['vcodec'] = 'none'
  72                             fdct['ext'] = 'm4a'
  73                         else:
  74                             fdct['tbr'] = int(mi.group(1))
  75
  76             formats.append(fdct)
  77
  78         self._sort_formats(formats)
  79
  80         thumbnails = sorted([((int(t.attrib['height']),int(t.attrib['width'])), t.text) for t in info.findall('images/image')])
  81         thumbs_dict = [{'resolution': res, 'url': t_url} for (res, t_url) in thumbnails]
  82
  83         metas_el = info.find('metas')
  84         upload_date = (
  85             metas_el.attrib.get('version') if metas_el is not None else None)
  86
  87         duration_el = info.find('length')
  88         duration = parse_duration(duration_el.text)
  89
  90         return {
  91             'id': info.attrib['id'],
  92             'title': info.find('headline').text,
  93             'formats': formats,
  94             'thumbnail': thumbnails[-1][1],
  95             'thumbnails': thumbs_dict,
  96             'description': info.find('description').text,
  97             'duration': duration,
  98             'upload_date': upload_date,
  99         }