_ Git - youtube-dl/blob - youtube_dl/extractor/cbsnews.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6
   7 from .common import InfoExtractor
   8 from ..utils import remove_start
   9
  10
  11 class CBSNewsIE(InfoExtractor):
  12     IE_DESC = 'CBS News'
  13     _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:[^/]+/)+(?P<id>[\da-z_-]+)'
  14
  15     _TESTS = [
  16         {
  17             'url': 'http://www.cbsnews.com/news/tesla-and-spacex-elon-musks-industrial-empire/',
  18             'info_dict': {
  19                 'id': 'tesla-and-spacex-elon-musks-industrial-empire',
  20                 'ext': 'flv',
  21                 'title': 'Tesla and SpaceX: Elon Musk\'s industrial empire',
  22                 'thumbnail': 'http://beta.img.cbsnews.com/i/2014/03/30/60147937-2f53-4565-ad64-1bdd6eb64679/60-0330-pelley-640x360.jpg',
  23                 'duration': 791,
  24             },
  25             'params': {
  26                 # rtmp download
  27                 'skip_download': True,
  28             },
  29         },
  30         {
  31             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
  32             'info_dict': {
  33                 'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
  34                 'ext': 'flv',
  35                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
  36                 'thumbnail': 're:^https?://.*\.jpg$',
  37                 'duration': 205,
  38                 'subtitles': {
  39                     'en': [{
  40                         'ext': 'ttml',
  41                     }],
  42                 },
  43             },
  44             'params': {
  45                 # rtmp download
  46                 'skip_download': True,
  47             },
  48         },
  49     ]
  50
  51     def _real_extract(self, url):
  52         mobj = re.match(self._VALID_URL, url)
  53         video_id = mobj.group('id')
  54
  55         webpage = self._download_webpage(url, video_id)
  56
  57         video_info = json.loads(self._html_search_regex(
  58             r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
  59             webpage, 'video JSON info'))
  60
  61         item = video_info['item'] if 'item' in video_info else video_info
  62         title = item.get('articleTitle') or item.get('hed')
  63         duration = item.get('duration')
  64         thumbnail = item.get('mediaImage') or item.get('thumbnail')
  65
  66         formats = []
  67         for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
  68             uri = item.get('media' + format_id + 'URI')
  69             if not uri:
  70                 continue
  71             uri = remove_start(uri, '{manifest:none}')
  72             fmt = {
  73                 'url': uri,
  74                 'format_id': format_id,
  75             }
  76             if uri.startswith('rtmp'):
  77                 play_path = re.sub(
  78                     r'{slistFilePath}', '',
  79                     uri.split('<break>')[-1].split('{break}')[-1])
  80                 play_path = re.sub(
  81                     r'{manifest:.+}.*$', '', play_path)
  82                 fmt.update({
  83                     'app': 'ondemand?auth=cbs',
  84                     'play_path': 'mp4:' + play_path,
  85                     'player_url': 'http://www.cbsnews.com/[[IMPORT]]/vidtech.cbsinteractive.com/player/3_3_0/CBSI_PLAYER_HD.swf',
  86                     'page_url': 'http://www.cbsnews.com',
  87                     'ext': 'flv',
  88                 })
  89             elif uri.endswith('.m3u8'):
  90                 fmt['ext'] = 'mp4'
  91             formats.append(fmt)
  92
  93         subtitles = {}
  94         if 'mpxRefId' in video_info:
  95             subtitles['en'] = [{
  96                 'ext': 'ttml',
  97                 'url': 'http://www.cbsnews.com/videos/captions/%s.adb_xml' % video_info['mpxRefId'],
  98             }]
  99
 100         return {
 101             'id': video_id,
 102             'title': title,
 103             'thumbnail': thumbnail,
 104             'duration': duration,
 105             'formats': formats,
 106             'subtitles': subtitles,
 107         }