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'http://(?:www\.)?kusi\.com/(?P<path>story/.+|video\?clipId=(?P<clipId>\d+))'
  20     _TESTS = [{
  21         'url': 'http://www.kusi.com/story/31183873/turko-files-case-closed-put-on-hold',
  22         'md5': 'f926e7684294cf8cb7bdf8858e1b3988',
  23         'info_dict': {
  24             'id': '12203019',
  25             'ext': 'mp4',
  26             'title': 'Turko Files: Case Closed! & Put On Hold!',
  27             'duration': 231,
  28         },
  29     }, {
  30         'url': 'http://kusi.com/video?clipId=12203019',
  31         'info_dict': {
  32             'id': '12203019',
  33             'ext': 'mp4',
  34             'title': 'Turko Files: Case Closed! & Put On Hold!',
  35             'duration': 231.0,
  36             'upload_date': '20160210',
  37             'timestamp': 1455087571,
  38             'thumbnail': 're:^https?://.*\.jpg$'
  39         },
  40         'params': {
  41             'skip_download': True,  # Same as previous one
  42         },
  43     }]
  44
  45     def _real_extract(self, url):
  46         mobj = re.match(self._VALID_URL, url)
  47         clip_id = mobj.group('clipId')
  48         video_id = clip_id or mobj.group('path')
  49
  50         webpage = self._download_webpage(url, video_id)
  51
  52         if clip_id is None:
  53             video_id = clip_id = self._html_search_regex(
  54                 r'"clipId"\s*,\s*"(\d+)"', webpage, 'clip id')
  55
  56         affiliate_id = self._search_regex(
  57             r'affiliateId\s*:\s*\'([^\']+)\'', webpage, 'affiliate id')
  58
  59         # See __Packages/worldnow/model/GalleryModel.as of WNGallery.swf
  60         xml_url = update_url_query('http://www.kusi.com/build.asp', {
  61             'buildtype': 'buildfeaturexmlrequest',
  62             'featureType': 'Clip',
  63             'featureid': clip_id,
  64             'affiliateno': affiliate_id,
  65             'clientgroupid': '1',
  66             'rnd': int(round(random.random() * 1000000)),
  67         })
  68
  69         doc = self._download_xml(xml_url, video_id)
  70
  71         video_title = xpath_text(doc, 'HEADLINE')
  72         duration = float_or_none(
  73             xpath_text(doc, 'DURATION', fatal=False), scale=1000)
  74         description = xpath_text(doc, 'ABSTRACT', fatal=False)
  75         thumbnail = xpath_text(doc, './THUMBNAILIMAGE/FILENAME', fatal=False)
  76         createtion_time = timeconvert(
  77             xpath_text(doc, 'rfc822creationdate', fatal=False))
  78
  79         quality_options = doc.find('{http://search.yahoo.com/mrss/}group').findall('{http://search.yahoo.com/mrss/}content')
  80         formats = []
  81         for quality in quality_options:
  82             formats.append({
  83                 'url': compat_urllib_parse_unquote_plus(quality.attrib['url']),
  84                 'height': int_or_none(quality.attrib.get('height')),
  85                 'width': int_or_none(quality.attrib.get('width')),
  86                 'vbr': float_or_none(quality.attrib.get('bitratebits'), scale=1024),
  87             })
  88         self._sort_formats(formats)
  89
  90         return {
  91             'id': video_id,
  92             'title': video_title,
  93             'description': description,
  94             'duration': duration,
  95             'formats': formats,
  96             'thumbnail': thumbnail,
  97             'timestamp': createtion_time,
  98         }