yt_dlp/extractor/howstuffworks.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     find_xpath_attr,
   4     int_or_none,
   5     js_to_json,
   6     unescapeHTML,
   7     determine_ext,
   8 )
   9
  10
  11 class HowStuffWorksIE(InfoExtractor):
  12     _VALID_URL = r'https?://[\da-z-]+\.(?:howstuffworks|stuff(?:(?:youshould|theydontwantyouto)know|toblowyourmind|momnevertoldyou)|(?:brain|car)stuffshow|fwthinking|geniusstuff)\.com/(?:[^/]+/)*(?:\d+-)?(?P<id>.+?)-video\.htm'
  13     _TESTS = [
  14         {
  15             'url': 'http://www.stufftoblowyourmind.com/videos/optical-illusions-video.htm',
  16             'md5': '76646a5acc0c92bf7cd66751ca5db94d',
  17             'info_dict': {
  18                 'id': '855410',
  19                 'ext': 'mp4',
  20                 'title': 'Your Trickster Brain: Optical Illusions -- Science on the Web',
  21                 'description': 'md5:e374ff9561f6833ad076a8cc0a5ab2fb',
  22             },
  23         },
  24         {
  25             'url': 'http://shows.howstuffworks.com/more-shows/why-does-balloon-stick-to-hair-video.htm',
  26             'only_matching': True,
  27         }
  28     ]
  29
  30     def _real_extract(self, url):
  31         display_id = self._match_id(url)
  32         webpage = self._download_webpage(url, display_id)
  33         clip_js = self._search_regex(
  34             r'(?s)var clip = ({.*?});', webpage, 'clip info')
  35         clip_info = self._parse_json(
  36             clip_js, display_id, transform_source=js_to_json)
  37
  38         video_id = clip_info['content_id']
  39         formats = []
  40         m3u8_url = clip_info.get('m3u8')
  41         if m3u8_url and determine_ext(m3u8_url) == 'm3u8':
  42             formats.extend(self._extract_m3u8_formats(m3u8_url, video_id, 'mp4', format_id='hls', fatal=True))
  43         flv_url = clip_info.get('flv_url')
  44         if flv_url:
  45             formats.append({
  46                 'url': flv_url,
  47                 'format_id': 'flv',
  48             })
  49         for video in clip_info.get('mp4', []):
  50             formats.append({
  51                 'url': video['src'],
  52                 'format_id': 'mp4-%s' % video['bitrate'],
  53                 'vbr': int_or_none(video['bitrate'].rstrip('k')),
  54             })
  55
  56         if not formats:
  57             smil = self._download_xml(
  58                 'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % video_id,
  59                 video_id, 'Downloading video SMIL')
  60
  61             http_base = find_xpath_attr(
  62                 smil,
  63                 './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'),
  64                 'name',
  65                 'httpBase').get('content')
  66
  67             URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=A&g=A'
  68
  69             for video in smil.findall(
  70                     './{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')):
  71                 vbr = int_or_none(video.attrib['system-bitrate'], scale=1000)
  72                 formats.append({
  73                     'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
  74                     'format_id': '%dk' % vbr,
  75                     'vbr': vbr,
  76                 })
  77
  78         return {
  79             'id': '%s' % video_id,
  80             'display_id': display_id,
  81             'title': unescapeHTML(clip_info['clip_title']),
  82             'description': unescapeHTML(clip_info.get('caption')),
  83             'thumbnail': clip_info.get('video_still_url'),
  84             'duration': int_or_none(clip_info.get('duration')),
  85             'formats': formats,
  86         }