_ 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         'md5': '041233212a0d06b179c87cbcca1577b8',
  19         'info_dict': {
  20             'id': '56f4ea68-bd21-4852-b08c-4de5b8354c60',
  21             'ext': 'mp4',
  22             'title': 'Hands-on with Microsoft Windows 8.1 Update',
  23             'description': 'The new update to the Windows 8 OS brings improved performance for mouse and keyboard users.',
  24             'thumbnail': 're:^http://.*/flmswindows8.jpg$',
  25             'uploader_id': 'sarah.mitroff@cbsinteractive.com',
  26             'uploader': 'Sarah Mitroff',
  27         }
  28     }
  29
  30     def _real_extract(self, url):
  31         mobj = re.match(self._VALID_URL, url)
  32         display_id = mobj.group('id')
  33
  34         webpage = self._download_webpage(url, display_id)
  35         data_json = self._html_search_regex(
  36             r"<div class=\"cnetVideoPlayer\"\s+.*?data-cnet-video-options='([^']+)'",
  37             webpage, 'data json')
  38         data = json.loads(data_json)
  39         vdata = data['video']
  40         if not vdata:
  41             vdata = data['videos'][0]
  42         if not vdata:
  43             raise ExtractorError('Cannot find video data')
  44
  45         video_id = vdata['id']
  46         title = vdata['headline']
  47         description = vdata.get('dek')
  48         thumbnail = vdata.get('image', {}).get('path')
  49         author = vdata.get('author')
  50         if author:
  51             uploader = '%s %s' % (author['firstName'], author['lastName'])
  52             uploader_id = author.get('email')
  53         else:
  54             uploader = None
  55             uploader_id = None
  56
  57         formats = [{
  58             'format_id': '%s-%s-%s' % (
  59                 f['type'], f['format'],
  60                 int_or_none(f.get('bitrate'), 1000, default='')),
  61             'url': f['uri'],
  62             'tbr': int_or_none(f.get('bitrate'), 1000),
  63         } for f in vdata['files']['data']]
  64         self._sort_formats(formats)
  65
  66         return {
  67             'id': video_id,
  68             'display_id': display_id,
  69             'title': title,
  70             'formats': formats,
  71             'description': description,
  72             'uploader': uploader,
  73             'uploader_id': uploader_id,
  74             'thumbnail': thumbnail,
  75         }