_ 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?://(?:(?P<sub_domain>edition|www|money)\.)?cnn\.com/(?:video/(?:data/.+?|\?)/)?videos?/
  15         (?P<path>.+?/(?P<title>[^/]+?)(?:\.(?:[a-z\-]+)|(?=&)))'''
  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         '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         'md5': 'b5cc60c60a3477d185af8f19a2a26f4e',
  31         'info_dict': {
  32             'id': 'us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology',
  33             'ext': 'mp4',
  34             'title': "Student's epic speech stuns new freshmen",
  35             'description': "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"",
  36             'upload_date': '20130821',
  37         }
  38     }, {
  39         'url': 'http://www.cnn.com/video/data/2.0/video/living/2014/12/22/growing-america-nashville-salemtown-board-episode-1.hln.html',
  40         'md5': 'f14d02ebd264df951feb2400e2c25a1b',
  41         'info_dict': {
  42             'id': 'living/2014/12/22/growing-america-nashville-salemtown-board-episode-1.hln',
  43             'ext': 'mp4',
  44             'title': 'Nashville Ep. 1: Hand crafted skateboards',
  45             'description': 'md5:e7223a503315c9f150acac52e76de086',
  46             'upload_date': '20141222',
  47         }
  48     }, {
  49         'url': 'http://money.cnn.com/video/news/2016/08/19/netflix-stunning-stats.cnnmoney/index.html',
  50         'md5': '52a515dc1b0f001cd82e4ceda32be9d1',
  51         'info_dict': {
  52             'id': '/video/news/2016/08/19/netflix-stunning-stats.cnnmoney',
  53             'ext': 'mp4',
  54             'title': '5 stunning stats about Netflix',
  55             'description': 'Did you know that Netflix has more than 80 million members? Here are five facts about the online video distributor that you probably didn\'t know.',
  56             'upload_date': '20160819',
  57         }
  58     }, {
  59         'url': 'http://cnn.com/video/?/video/politics/2015/03/27/pkg-arizona-senator-church-attendance-mandatory.ktvk',
  60         'only_matching': True,
  61     }, {
  62         'url': 'http://cnn.com/video/?/video/us/2015/04/06/dnt-baker-refuses-anti-gay-order.wkmg',
  63         'only_matching': True,
  64     }, {
  65         'url': 'http://edition.cnn.com/videos/arts/2016/04/21/olympic-games-cultural-a-z-brazil.cnn',
  66         'only_matching': True,
  67     }]
  68
  69     _CONFIG = {
  70         # http://edition.cnn.com/.element/apps/cvp/3.0/cfg/spider/cnn/expansion/config.xml
  71         'edition': {
  72             'data_src': 'http://edition.cnn.com/video/data/3.0/video/%s/index.xml',
  73             'media_src': 'http://pmd.cdn.turner.com/cnn/big',
  74         },
  75         # http://money.cnn.com/.element/apps/cvp2/cfg/config.xml
  76         'money': {
  77             'data_src': 'http://money.cnn.com/video/data/4.0/video/%s.xml',
  78             'media_src': 'http://ht3.cdn.turner.com/money/big',
  79         },
  80     }
  81
  82     def _real_extract(self, url):
  83         sub_domain, path, page_title = re.match(self._VALID_URL, url).groups()
  84         if sub_domain not in ('money', 'edition'):
  85             sub_domain = 'edition'
  86         config = self._CONFIG[sub_domain]
  87         info_url = config['data_src'] % path
  88         info = self._download_xml(info_url, page_title)
  89
  90         formats = []
  91         rex = re.compile(r'''(?x)
  92             (?P<width>[0-9]+)x(?P<height>[0-9]+)
  93             (?:_(?P<bitrate>[0-9]+)k)?
  94         ''')
  95         for f in info.findall('files/file'):
  96             video_url = config['media_src'] + f.text.strip()
  97             fdct = {
  98                 'format_id': f.attrib['bitrate'],
  99                 'url': video_url,
 100             }
 101
 102             mf = rex.match(f.attrib['bitrate'])
 103             if mf:
 104                 fdct['width'] = int(mf.group('width'))
 105                 fdct['height'] = int(mf.group('height'))
 106                 fdct['tbr'] = int_or_none(mf.group('bitrate'))
 107             else:
 108                 mf = rex.search(f.text)
 109                 if mf:
 110                     fdct['width'] = int(mf.group('width'))
 111                     fdct['height'] = int(mf.group('height'))
 112                     fdct['tbr'] = int_or_none(mf.group('bitrate'))
 113                 else:
 114                     mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
 115                     if mi:
 116                         if mi.group(1) == 'audio':
 117                             fdct['vcodec'] = 'none'
 118                             fdct['ext'] = 'm4a'
 119                         else:
 120                             fdct['tbr'] = int(mi.group(1))
 121
 122             formats.append(fdct)
 123
 124         self._sort_formats(formats)
 125
 126         thumbnails = [{
 127             'height': int(t.attrib['height']),
 128             'width': int(t.attrib['width']),
 129             'url': t.text,
 130         } for t in info.findall('images/image')]
 131
 132         metas_el = info.find('metas')
 133         upload_date = (
 134             metas_el.attrib.get('version') if metas_el is not None else None)
 135
 136         duration_el = info.find('length')
 137         duration = parse_duration(duration_el.text)
 138
 139         return {
 140             'id': info.attrib['id'],
 141             'title': info.find('headline').text,
 142             'formats': formats,
 143             'thumbnails': thumbnails,
 144             'description': info.find('description').text,
 145             'duration': duration,
 146             'upload_date': upload_date,
 147         }
 148
 149
 150 class CNNBlogsIE(InfoExtractor):
 151     _VALID_URL = r'https?://[^\.]+\.blogs\.cnn\.com/.+'
 152     _TEST = {
 153         'url': 'http://reliablesources.blogs.cnn.com/2014/02/09/criminalizing-journalism/',
 154         'md5': '3e56f97b0b6ffb4b79f4ea0749551084',
 155         'info_dict': {
 156             'id': 'bestoftv/2014/02/09/criminalizing-journalism.cnn',
 157             'ext': 'mp4',
 158             'title': 'Criminalizing journalism?',
 159             'description': 'Glenn Greenwald responds to comments made this week on Capitol Hill that journalists could be criminal accessories.',
 160             'upload_date': '20140209',
 161         },
 162         'add_ie': ['CNN'],
 163     }
 164
 165     def _real_extract(self, url):
 166         webpage = self._download_webpage(url, url_basename(url))
 167         cnn_url = self._html_search_regex(r'data-url="(.+?)"', webpage, 'cnn url')
 168         return {
 169             '_type': 'url',
 170             'url': cnn_url,
 171             'ie_key': CNNIE.ie_key(),
 172         }
 173
 174
 175 class CNNArticleIE(InfoExtractor):
 176     _VALID_URL = r'https?://(?:(?:edition|www)\.)?cnn\.com/(?!videos?/)'
 177     _TEST = {
 178         'url': 'http://www.cnn.com/2014/12/21/politics/obama-north-koreas-hack-not-war-but-cyber-vandalism/',
 179         'md5': '689034c2a3d9c6dc4aa72d65a81efd01',
 180         'info_dict': {
 181             'id': 'bestoftv/2014/12/21/ip-north-korea-obama.cnn',
 182             'ext': 'mp4',
 183             'title': 'Obama: Cyberattack not an act of war',
 184             'description': 'md5:51ce6750450603795cad0cdfbd7d05c5',
 185             'upload_date': '20141221',
 186         },
 187         'add_ie': ['CNN'],
 188     }
 189
 190     def _real_extract(self, url):
 191         webpage = self._download_webpage(url, url_basename(url))
 192         cnn_url = self._html_search_regex(r"video:\s*'([^']+)'", webpage, 'cnn url')
 193         return {
 194             '_type': 'url',
 195             'url': 'http://cnn.com/video/?/video/' + cnn_url,
 196             'ie_key': CNNIE.ie_key(),
 197         }