Raphaël G. Git Repositories - youtubedl/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
   9
  10 class CBSNewsIE(InfoExtractor):
  11     IE_DESC = 'CBS News'
  12     _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:[^/]+/)+(?P<id>[\da-z_-]+)'
  13
  14     _TESTS = [
  15         {
  16             'url': 'http://www.cbsnews.com/news/tesla-and-spacex-elon-musks-industrial-empire/',
  17             'info_dict': {
  18                 'id': 'tesla-and-spacex-elon-musks-industrial-empire',
  19                 'ext': 'flv',
  20                 'title': 'Tesla and SpaceX: Elon Musk\'s industrial empire',
  21                 'thumbnail': 'http://beta.img.cbsnews.com/i/2014/03/30/60147937-2f53-4565-ad64-1bdd6eb64679/60-0330-pelley-640x360.jpg',
  22                 'duration': 791,
  23             },
  24             'params': {
  25                 # rtmp download
  26                 'skip_download': True,
  27             },
  28         },
  29         {
  30             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
  31             'info_dict': {
  32                 'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
  33                 'ext': 'flv',
  34                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
  35                 'thumbnail': 're:^https?://.*\.jpg$',
  36                 'duration': 205,
  37             },
  38             'params': {
  39                 # rtmp download
  40                 'skip_download': True,
  41             },
  42         },
  43     ]
  44
  45     def _real_extract(self, url):
  46         mobj = re.match(self._VALID_URL, url)
  47         video_id = mobj.group('id')
  48
  49         webpage = self._download_webpage(url, video_id)
  50
  51         video_info = json.loads(self._html_search_regex(
  52             r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
  53             webpage, 'video JSON info'))
  54
  55         item = video_info['item'] if 'item' in video_info else video_info
  56         title = item.get('articleTitle') or item.get('hed')
  57         duration = item.get('duration')
  58         thumbnail = item.get('mediaImage') or item.get('thumbnail')
  59
  60         formats = []
  61         for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
  62             uri = item.get('media' + format_id + 'URI')
  63             if not uri:
  64                 continue
  65             fmt = {
  66                 'url': uri,
  67                 'format_id': format_id,
  68             }
  69             if uri.startswith('rtmp'):
  70                 fmt.update({
  71                     'app': 'ondemand?auth=cbs',
  72                     'play_path': 'mp4:' + uri.split('<break>')[-1],
  73                     'player_url': 'http://www.cbsnews.com/[[IMPORT]]/vidtech.cbsinteractive.com/player/3_3_0/CBSI_PLAYER_HD.swf',
  74                     'page_url': 'http://www.cbsnews.com',
  75                     'ext': 'flv',
  76                 })
  77             elif uri.endswith('.m3u8'):
  78                 fmt['ext'] = 'mp4'
  79             formats.append(fmt)
  80
  81         return {
  82             'id': video_id,
  83             'title': title,
  84             'thumbnail': thumbnail,
  85             'duration': duration,
  86             'formats': formats,
  87         }