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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11 )
  12
  13
  14 class CNETIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?cnet\.com/videos/(?P<id>[^/]+)/'
  16     _TEST = {
  17         'url': 'http://www.cnet.com/videos/hands-on-with-microsofts-windows-8-1-update/',
  18         'info_dict': {
  19             'id': '56f4ea68-bd21-4852-b08c-4de5b8354c60',
  20             'ext': 'flv',
  21             'title': 'Hands-on with Microsoft Windows 8.1 Update',
  22             'description': 'The new update to the Windows 8 OS brings improved performance for mouse and keyboard users.',
  23             'thumbnail': 're:^http://.*/flmswindows8.jpg$',
  24             'uploader_id': '6085384d-619e-11e3-b231-14feb5ca9861',
  25             'uploader': 'Sarah Mitroff',
  26         },
  27         'params': {
  28             'skip_download': 'requires rtmpdump',
  29         }
  30     }
  31
  32     def _real_extract(self, url):
  33         display_id = self._match_id(url)
  34         webpage = self._download_webpage(url, display_id)
  35
  36         data_json = self._html_search_regex(
  37             r"<div class=\"cnetVideoPlayer\"\s+.*?data-cnet-video-options='([^']+)'",
  38             webpage, 'data json')
  39         data = json.loads(data_json)
  40         vdata = data['video']
  41         if not vdata:
  42             vdata = data['videos'][0]
  43         if not vdata:
  44             raise ExtractorError('Cannot find video data')
  45
  46         mpx_account = data['config']['players']['default']['mpx_account']
  47         vid = vdata['files']['rtmp']
  48         tp_link = 'http://link.theplatform.com/s/%s/%s' % (mpx_account, vid)
  49
  50         video_id = vdata['id']
  51         title = vdata.get('headline')
  52         if title is None:
  53             title = vdata.get('title')
  54         if title is None:
  55             raise ExtractorError('Cannot find title!')
  56         thumbnail = vdata.get('image', {}).get('path')
  57         author = vdata.get('author')
  58         if author:
  59             uploader = '%s %s' % (author['firstName'], author['lastName'])
  60             uploader_id = author.get('id')
  61         else:
  62             uploader = None
  63             uploader_id = None
  64
  65         return {
  66             '_type': 'url_transparent',
  67             'url': tp_link,
  68             'id': video_id,
  69             'display_id': display_id,
  70             'title': title,
  71             'uploader': uploader,
  72             'uploader_id': uploader_id,
  73             'thumbnail': thumbnail,
  74         }