youtube_dl/extractor/howstuffworks.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import find_xpath_attr
   5
   6
   7 class HowStuffWorksIE(InfoExtractor):
   8     _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P<id>.+?)-video\.htm'
   9     _TESTS = [
  10         {
  11             'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm',
  12             'info_dict': {
  13                 'id': '450221',
  14                 'ext': 'flv',
  15                 'title': 'Cool Jobs - Iditarod Musher',
  16                 'description': 'Cold sleds, freezing temps and warm dog breath... an Iditarod musher\'s dream. Kasey-Dee Gardner jumps on a sled to find out what the big deal is.',
  17                 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/5266.jpg',
  18             },
  19         },
  20         {
  21             'url': 'http://adventure.howstuffworks.com/7199-survival-zone-food-and-water-in-the-savanna-video.htm',
  22             'info_dict': {
  23                 'id': '453464',
  24                 'ext': 'mp4',
  25                 'title': 'Survival Zone: Food and Water In the Savanna',
  26                 'description': 'Learn how to find both food and water while trekking in the African savannah. In this video from the Discovery Channel.',
  27                 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/7199.jpg',
  28             },
  29         },
  30         {
  31             'url': 'http://entertainment.howstuffworks.com/arts/2706-sword-swallowing-1-by-dan-meyer-video.htm',
  32             'info_dict': {
  33                 'id': '440011',
  34                 'ext': 'flv',
  35                 'title': 'Sword Swallowing #1 by Dan Meyer',
  36                 'description': 'Video footage (1 of 3) used by permission of the owner Dan Meyer through Sword Swallowers Association International <www.swordswallow.org>',
  37                 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/118306353233.jpg',
  38             },
  39         },
  40     ]
  41
  42     def _real_extract(self, url):
  43         display_id = self._match_id(url)
  44         webpage = self._download_webpage(url, display_id)
  45         clip_info = self._search_regex('(?s)var clip = {(.*?)};', webpage, 'clip info')
  46
  47         def extract_clip_info(key, clip_info, name=None, **kargs):
  48             if name is None:
  49                 name = key
  50             return self._html_search_regex(
  51                 r"\s*%s\s*: '?([^'\n]*[^'\n,])" % key, clip_info, name, **kargs)
  52
  53         video_id = extract_clip_info('content_id', clip_info, 'video id')
  54         formats = []
  55         m3u8_url = extract_clip_info('m3u8', clip_info, 'm3u8 url', default=None)
  56         if m3u8_url is not None:
  57             formats += self._extract_m3u8_formats(m3u8_url , video_id, 'mp4')
  58         mp4 = self._parse_json(
  59             extract_clip_info(
  60                 'mp4', clip_info, 'formats').replace('},]','}]'), video_id)
  61         for video in mp4:
  62             formats.append({
  63                 'url': video['src'],
  64                 'format_id': video['bitrate'],
  65                 'vbr': int(video['bitrate'].rstrip('k')),
  66             })
  67
  68         if not formats:
  69             smil = self._download_xml(
  70                 'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % video_id,
  71                 video_id, 'Downloading video SMIL')
  72
  73             http_base = find_xpath_attr(
  74                 smil,
  75                 './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'),
  76                 'name',
  77                 'httpBase').get('content')
  78
  79             URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=A&g=A'
  80
  81             for video in smil.findall(
  82                     './{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')):
  83                 vbr = int(video.attrib['system-bitrate']) / 1000
  84                 formats.append({
  85                     'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
  86                     'format_id': '%dk' % vbr,
  87                     'vbr': vbr,
  88                 })
  89
  90         self._sort_formats(formats)
  91
  92         return {
  93             'id': video_id,
  94             'display_id': display_id,
  95             'title': extract_clip_info('clip_title', clip_info, 'title'),
  96             'description': extract_clip_info('caption', clip_info, 'description', fatal=False),
  97             'thumbnail': extract_clip_info('video_still_url', clip_info, 'thumbnail'),
  98             'duration': extract_clip_info('duration', clip_info),
  99             'formats': formats,
 100         }