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>[^/]+?)(?:\.(?:[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://cnn.com/video/?/video/politics/2015/03/27/pkg-arizona-senator-church-attendance-mandatory.ktvk',
  50         'only_matching': True,
  51     }, {
  52         'url': 'http://cnn.com/video/?/video/us/2015/04/06/dnt-baker-refuses-anti-gay-order.wkmg',
  53         'only_matching': True,
  54     }]
  55
  56     def _real_extract(self, url):
  57         mobj = re.match(self._VALID_URL, url)
  58         path = mobj.group('path')
  59         page_title = mobj.group('title')
  60         info_url = 'http://edition.cnn.com/video/data/3.0/%s/index.xml' % path
  61         info = self._download_xml(info_url, page_title)
  62
  63         formats = []
  64         rex = re.compile(r'''(?x)
  65             (?P<width>[0-9]+)x(?P<height>[0-9]+)
  66             (?:_(?P<bitrate>[0-9]+)k)?
  67         ''')
  68         for f in info.findall('files/file'):
  69             video_url = 'http://ht.cdn.turner.com/cnn/big%s' % (f.text.strip())
  70             fdct = {
  71                 'format_id': f.attrib['bitrate'],
  72                 'url': video_url,
  73             }
  74
  75             mf = rex.match(f.attrib['bitrate'])
  76             if mf:
  77                 fdct['width'] = int(mf.group('width'))
  78                 fdct['height'] = int(mf.group('height'))
  79                 fdct['tbr'] = int_or_none(mf.group('bitrate'))
  80             else:
  81                 mf = rex.search(f.text)
  82                 if mf:
  83                     fdct['width'] = int(mf.group('width'))
  84                     fdct['height'] = int(mf.group('height'))
  85                     fdct['tbr'] = int_or_none(mf.group('bitrate'))
  86                 else:
  87                     mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
  88                     if mi:
  89                         if mi.group(1) == 'audio':
  90                             fdct['vcodec'] = 'none'
  91                             fdct['ext'] = 'm4a'
  92                         else:
  93                             fdct['tbr'] = int(mi.group(1))
  94
  95             formats.append(fdct)
  96
  97         self._sort_formats(formats)
  98
  99         thumbnails = [{
 100             'height': int(t.attrib['height']),
 101             'width': int(t.attrib['width']),
 102             'url': t.text,
 103         } for t in info.findall('images/image')]
 104
 105         metas_el = info.find('metas')
 106         upload_date = (
 107             metas_el.attrib.get('version') if metas_el is not None else None)
 108
 109         duration_el = info.find('length')
 110         duration = parse_duration(duration_el.text)
 111
 112         return {
 113             'id': info.attrib['id'],
 114             'title': info.find('headline').text,
 115             'formats': formats,
 116             'thumbnails': thumbnails,
 117             'description': info.find('description').text,
 118             'duration': duration,
 119             'upload_date': upload_date,
 120         }
 121
 122
 123 class CNNBlogsIE(InfoExtractor):
 124     _VALID_URL = r'https?://[^\.]+\.blogs\.cnn\.com/.+'
 125     _TEST = {
 126         'url': 'http://reliablesources.blogs.cnn.com/2014/02/09/criminalizing-journalism/',
 127         'md5': '3e56f97b0b6ffb4b79f4ea0749551084',
 128         'info_dict': {
 129             'id': 'bestoftv/2014/02/09/criminalizing-journalism.cnn',
 130             'ext': 'mp4',
 131             'title': 'Criminalizing journalism?',
 132             'description': 'Glenn Greenwald responds to comments made this week on Capitol Hill that journalists could be criminal accessories.',
 133             'upload_date': '20140209',
 134         },
 135         'add_ie': ['CNN'],
 136     }
 137
 138     def _real_extract(self, url):
 139         webpage = self._download_webpage(url, url_basename(url))
 140         cnn_url = self._html_search_regex(r'data-url="(.+?)"', webpage, 'cnn url')
 141         return {
 142             '_type': 'url',
 143             'url': cnn_url,
 144             'ie_key': CNNIE.ie_key(),
 145         }
 146
 147
 148 class CNNArticleIE(InfoExtractor):
 149     _VALID_URL = r'https?://(?:(?:edition|www)\.)?cnn\.com/(?!video/)'
 150     _TEST = {
 151         'url': 'http://www.cnn.com/2014/12/21/politics/obama-north-koreas-hack-not-war-but-cyber-vandalism/',
 152         'md5': '689034c2a3d9c6dc4aa72d65a81efd01',
 153         'info_dict': {
 154             'id': 'bestoftv/2014/12/21/ip-north-korea-obama.cnn',
 155             'ext': 'mp4',
 156             'title': 'Obama: Cyberattack not an act of war',
 157             'description': 'md5:51ce6750450603795cad0cdfbd7d05c5',
 158             'upload_date': '20141221',
 159         },
 160         'add_ie': ['CNN'],
 161     }
 162
 163     def _real_extract(self, url):
 164         webpage = self._download_webpage(url, url_basename(url))
 165         cnn_url = self._html_search_regex(r"video:\s*'([^']+)'", webpage, 'cnn url')
 166         return {
 167             '_type': 'url',
 168             'url': 'http://cnn.com/video/?/video/' + cnn_url,
 169             'ie_key': CNNIE.ie_key(),
 170         }