youtube_dl/extractor/kusi.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..compat import compat_urllib_parse_unquote_plus
   9 from ..utils import (
  10     int_or_none,
  11     float_or_none,
  12     timeconvert,
  13     update_url_query,
  14     xpath_text,
  15 )
  16
  17
  18 class KUSIIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:www\.)?kusi\.com/(?P<path>story/.+|video\?clipId=(?P<clipId>\d+))'
  20     _TESTS = [{
  21         'url': 'http://www.kusi.com/story/32849881/turko-files-refused-to-help-it-aint-right',
  22         'md5': '4e76ce8e53660ce9697d06c0ba6fc47d',
  23         'info_dict': {
  24             'id': '12689020',
  25             'ext': 'mp4',
  26             'title': "Turko Files: Refused to Help, It Ain't Right!",
  27             'duration': 223.586,
  28             'upload_date': '20160826',
  29             'timestamp': 1472233118,
  30             'thumbnail': 're:^https?://.*\.jpg$'
  31         },
  32     }, {
  33         'url': 'http://kusi.com/video?clipId=12203019',
  34         'only_matching': True,
  35     }]
  36
  37     def _real_extract(self, url):
  38         mobj = re.match(self._VALID_URL, url)
  39         clip_id = mobj.group('clipId')
  40         video_id = clip_id or mobj.group('path')
  41
  42         webpage = self._download_webpage(url, video_id)
  43
  44         if clip_id is None:
  45             video_id = clip_id = self._html_search_regex(
  46                 r'"clipId"\s*,\s*"(\d+)"', webpage, 'clip id')
  47
  48         affiliate_id = self._search_regex(
  49             r'affiliateId\s*:\s*\'([^\']+)\'', webpage, 'affiliate id')
  50
  51         # See __Packages/worldnow/model/GalleryModel.as of WNGallery.swf
  52         xml_url = update_url_query('http://www.kusi.com/build.asp', {
  53             'buildtype': 'buildfeaturexmlrequest',
  54             'featureType': 'Clip',
  55             'featureid': clip_id,
  56             'affiliateno': affiliate_id,
  57             'clientgroupid': '1',
  58             'rnd': int(round(random.random() * 1000000)),
  59         })
  60
  61         doc = self._download_xml(xml_url, video_id)
  62
  63         video_title = xpath_text(doc, 'HEADLINE', fatal=True)
  64         duration = float_or_none(xpath_text(doc, 'DURATION'), scale=1000)
  65         description = xpath_text(doc, 'ABSTRACT')
  66         thumbnail = xpath_text(doc, './THUMBNAILIMAGE/FILENAME')
  67         createtion_time = timeconvert(xpath_text(doc, 'rfc822creationdate'))
  68
  69         quality_options = doc.find('{http://search.yahoo.com/mrss/}group').findall('{http://search.yahoo.com/mrss/}content')
  70         formats = []
  71         for quality in quality_options:
  72             formats.append({
  73                 'url': compat_urllib_parse_unquote_plus(quality.attrib['url']),
  74                 'height': int_or_none(quality.attrib.get('height')),
  75                 'width': int_or_none(quality.attrib.get('width')),
  76                 'vbr': float_or_none(quality.attrib.get('bitratebits'), scale=1000),
  77             })
  78         self._sort_formats(formats)
  79
  80         return {
  81             'id': video_id,
  82             'title': video_title,
  83             'description': description,
  84             'duration': duration,
  85             'formats': formats,
  86             'thumbnail': thumbnail,
  87             'timestamp': createtion_time,
  88         }