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     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|hln)(?:-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         "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
  50     def _real_extract(self, url):
  51         mobj = re.match(self._VALID_URL, url)
  52         path = mobj.group('path')
  53         page_title = mobj.group('title')
  54         info_url = 'http://cnn.com/video/data/3.0/%s/index.xml' % path
  55         info = self._download_xml(info_url, page_title)
  56
  57         formats = []
  58         rex = re.compile(r'''(?x)
  59             (?P<width>[0-9]+)x(?P<height>[0-9]+)
  60             (?:_(?P<bitrate>[0-9]+)k)?
  61         ''')
  62         for f in info.findall('files/file'):
  63             video_url = 'http://ht.cdn.turner.com/cnn/big%s' % (f.text.strip())
  64             fdct = {
  65                 'format_id': f.attrib['bitrate'],
  66                 'url': video_url,
  67             }
  68
  69             mf = rex.match(f.attrib['bitrate'])
  70             if mf:
  71                 fdct['width'] = int(mf.group('width'))
  72                 fdct['height'] = int(mf.group('height'))
  73                 fdct['tbr'] = int_or_none(mf.group('bitrate'))
  74             else:
  75                 mf = rex.search(f.text)
  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                     mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
  82                     if mi:
  83                         if mi.group(1) == 'audio':
  84                             fdct['vcodec'] = 'none'
  85                             fdct['ext'] = 'm4a'
  86                         else:
  87                             fdct['tbr'] = int(mi.group(1))
  88
  89             formats.append(fdct)
  90
  91         self._sort_formats(formats)
  92
  93         thumbnails = [{
  94             'height': int(t.attrib['height']),
  95             'width': int(t.attrib['width']),
  96             'url': t.text,
  97         } for t in info.findall('images/image')]
  98
  99         metas_el = info.find('metas')
 100         upload_date = (
 101             metas_el.attrib.get('version') if metas_el is not None else None)
 102
 103         duration_el = info.find('length')
 104         duration = parse_duration(duration_el.text)
 105
 106         return {
 107             'id': info.attrib['id'],
 108             'title': info.find('headline').text,
 109             'formats': formats,
 110             'thumbnails': thumbnails,
 111             'description': info.find('description').text,
 112             'duration': duration,
 113             'upload_date': upload_date,
 114         }
 115
 116
 117 class CNNBlogsIE(InfoExtractor):
 118     _VALID_URL = r'https?://[^\.]+\.blogs\.cnn\.com/.+'
 119     _TEST = {
 120         'url': 'http://reliablesources.blogs.cnn.com/2014/02/09/criminalizing-journalism/',
 121         'md5': '3e56f97b0b6ffb4b79f4ea0749551084',
 122         'info_dict': {
 123             'id': 'bestoftv/2014/02/09/criminalizing-journalism.cnn',
 124             'ext': 'mp4',
 125             'title': 'Criminalizing journalism?',
 126             'description': 'Glenn Greenwald responds to comments made this week on Capitol Hill that journalists could be criminal accessories.',
 127             'upload_date': '20140209',
 128         },
 129         'add_ie': ['CNN'],
 130     }
 131
 132     def _real_extract(self, url):
 133         webpage = self._download_webpage(url, url_basename(url))
 134         cnn_url = self._html_search_regex(r'data-url="(.+?)"', webpage, 'cnn url')
 135         return {
 136             '_type': 'url',
 137             'url': cnn_url,
 138             'ie_key': CNNIE.ie_key(),
 139         }
 140
 141
 142 class CNNArticleIE(InfoExtractor):
 143     _VALID_URL = r'https?://(?:(?:edition|www)\.)?cnn\.com/(?!video/)'
 144     _TEST = {
 145         'url': 'http://www.cnn.com/2014/12/21/politics/obama-north-koreas-hack-not-war-but-cyber-vandalism/',
 146         'md5': '275b326f85d80dff7592a9820f5dc887',
 147         'info_dict': {
 148             'id': 'bestoftv/2014/12/21/sotu-crowley-president-obama-north-korea-not-going-to-be-intimidated.cnn',
 149             'ext': 'mp4',
 150             'title': 'Obama: We\'re not going to be intimidated',
 151             'description': 'md5:e735586f3dc936075fa654a4d91b21f9',
 152             'upload_date': '20141220',
 153         },
 154         'add_ie': ['CNN'],
 155     }
 156
 157     def _real_extract(self, url):
 158         webpage = self._download_webpage(url, url_basename(url))
 159         cnn_url = self._html_search_regex(r"video:\s*'([^']+)'", webpage, 'cnn url')
 160         return {
 161             '_type': 'url',
 162             'url': 'http://cnn.com/video/?/video/' + cnn_url,
 163             'ie_key': CNNIE.ie_key(),
 164         }