_ 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     int_or_none,
  10 )
  11
  12
  13 class CNETIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?cnet\.com/videos/(?P<id>[^/]+)/'
  15     _TEST = {
  16         'url': 'http://www.cnet.com/videos/hands-on-with-microsofts-windows-8-1-update/',
  17         'md5': '041233212a0d06b179c87cbcca1577b8',
  18         'info_dict': {
  19             'id': '56f4ea68-bd21-4852-b08c-4de5b8354c60',
  20             'ext': 'mp4',
  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': 'sarah.mitroff@cbsinteractive.com',
  25             'uploader': 'Sarah Mitroff',
  26         }
  27     }
  28
  29     def _real_extract(self, url):
  30         mobj = re.match(self._VALID_URL, url)
  31         display_id = mobj.group('id')
  32
  33         webpage = self._download_webpage(url, display_id)
  34         data_json = self._html_search_regex(
  35             r"<div class=\"cnetVideoPlayer\" data-cnet-video-options='([^']+)'",
  36             webpage, 'data json')
  37         data = json.loads(data_json)
  38         vdata = data['video']
  39
  40         video_id = vdata['id']
  41         title = vdata['headline']
  42         description = vdata.get('dek')
  43         thumbnail = vdata.get('image', {}).get('path')
  44         author = vdata.get('author')
  45         if author:
  46             uploader = '%s %s' % (author['firstName'], author['lastName'])
  47             uploader_id = author.get('email')
  48         else:
  49             uploader = None
  50             uploader_id = None
  51
  52         formats = [{
  53             'format_id': '%s-%s-%s' % (
  54                 f['type'], f['format'],
  55                 int_or_none(f.get('bitrate'), 1000, default='')),
  56             'url': f['uri'],
  57             'tbr': int_or_none(f.get('bitrate'), 1000),
  58         } for f in vdata['files']['data']]
  59         self._sort_formats(formats)
  60
  61         return {
  62             'id': video_id,
  63             'display_id': display_id,
  64             'title': title,
  65             'formats': formats,
  66             'description': description,
  67             'uploader': uploader,
  68             'uploader_id': uploader_id,
  69             'thumbnail': thumbnail,
  70         }