Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/cnet.py

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