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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urllib_parse_unquote_plus
   8 from ..utils import int_or_none
   9
  10
  11 class KUSIIE(InfoExtractor):
  12     _VALID_URL = r'http://(?:www\.)?kusi\.com/(?P<path>story/.+|video\?clipId=(?P<clipId>\d+))'
  13     _TESTS = [{
  14         'url': 'http://www.kusi.com/story/31183873/turko-files-case-closed-put-on-hold',
  15         'md5': 'f926e7684294cf8cb7bdf8858e1b3988',
  16         'info_dict': {
  17             'id': '12203019',
  18             'ext': 'mp4',
  19             'title': 'Turko Files: Case Closed! & Put On Hold!',
  20             'duration': 231000,
  21         },
  22     }, {
  23         'url': 'http://kusi.com/video?clipId=12203019',
  24         'info_dict': {
  25             'id': '12203019',
  26             'ext': 'mp4',
  27             'title': 'Turko Files: Case Closed! & Put On Hold!',
  28             'duration': 231000,
  29         },
  30         'params': {
  31             'skip_download': True,  # Same as previous one
  32         },
  33     }]
  34
  35     def _real_extract(self, url):
  36         mobj = re.match(self._VALID_URL, url)
  37
  38         if mobj.group('clipId') is not None:
  39             video_id = mobj.group('clipId')
  40         else:
  41             webpage = self._download_webpage(url, mobj.group('path'))
  42             video_id = self._html_search_regex(r'"clipId", "(\d+)"', webpage,
  43                                                'clipId')
  44
  45         xml_url = 'http://www.kusi.com/build.asp?buildtype=buildfeaturexml'\
  46                   'request&featureType=Clip&featureid={0}&affiliateno=956&'\
  47                   'clientgroupid=1&rnd=562461'.format(video_id)
  48         doc = self._download_xml(xml_url, video_id,
  49                                  note='Downloading video info',
  50                                  errnote='Failed to download video info')
  51
  52         video_title = doc.find('HEADLINE').text
  53         duration = int_or_none(doc.find('DURATION'), get_attr='text')
  54         description = doc.find('ABSTRACT')
  55
  56         quality_options = doc.find('{http://search.yahoo.com/mrss/}group').findall('{http://search.yahoo.com/mrss/}content')
  57         formats = []
  58         for quality in quality_options:
  59             if 'height' in quality.attrib:
  60                 formats.append({
  61                     'url': compat_urllib_parse_unquote_plus(quality.attrib['url']),
  62                     'height': quality.attrib['height'],
  63                 })
  64         self._sort_formats(formats)
  65
  66         return {
  67             'id': video_id,
  68             'title': video_title,
  69             'description': description,
  70             'duration': duration,
  71             'formats': formats,
  72         }