youtube_dl/extractor/howstuffworks.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     find_xpath_attr,
   6     int_or_none,
   7     js_to_json,
   8     unescapeHTML,
   9 )
  10
  11
  12 class HowStuffWorksIE(InfoExtractor):
  13     _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P<id>.+?)-video\.htm'
  14     _TESTS = [
  15         {
  16             'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm',
  17             'info_dict': {
  18                 'id': '450221',
  19                 'ext': 'flv',
  20                 'title': 'Cool Jobs - Iditarod Musher',
  21                 '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.',
  22                 'display_id': 'cool-jobs-iditarod-musher',
  23                 'thumbnail': 're:^https?://.*\.jpg$',
  24                 'duration': 161,
  25             },
  26         },
  27         {
  28             'url': 'http://adventure.howstuffworks.com/7199-survival-zone-food-and-water-in-the-savanna-video.htm',
  29             'info_dict': {
  30                 'id': '453464',
  31                 'ext': 'mp4',
  32                 'title': 'Survival Zone: Food and Water In the Savanna',
  33                 'description': 'Learn how to find both food and water while trekking in the African savannah. In this video from the Discovery Channel.',
  34                 'display_id': 'survival-zone-food-and-water-in-the-savanna',
  35                 'thumbnail': 're:^https?://.*\.jpg$',
  36             },
  37         },
  38         {
  39             'url': 'http://entertainment.howstuffworks.com/arts/2706-sword-swallowing-1-by-dan-meyer-video.htm',
  40             'info_dict': {
  41                 'id': '440011',
  42                 'ext': 'flv',
  43                 'title': 'Sword Swallowing #1 by Dan Meyer',
  44                 'description': 'Video footage (1 of 3) used by permission of the owner Dan Meyer through Sword Swallowers Association International <www.swordswallow.org>',
  45                 'display_id': 'sword-swallowing-1-by-dan-meyer',
  46                 'thumbnail': 're:^https?://.*\.jpg$',
  47             },
  48         },
  49     ]
  50
  51     def _real_extract(self, url):
  52         display_id = self._match_id(url)
  53         webpage = self._download_webpage(url, display_id)
  54         clip_js = self._search_regex(
  55             r'(?s)var clip = ({.*?});', webpage, 'clip info')
  56         clip_info = self._parse_json(
  57             clip_js, display_id, transform_source=js_to_json)
  58
  59         video_id = clip_info['content_id']
  60         formats = []
  61         m3u8_url = clip_info.get('m3u8')
  62         if m3u8_url:
  63             formats += self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
  64         for video in clip_info.get('mp4', []):
  65             formats.append({
  66                 'url': video['src'],
  67                 'format_id': video['bitrate'],
  68                 'vbr': int(video['bitrate'].rstrip('k')),
  69             })
  70
  71         if not formats:
  72             smil = self._download_xml(
  73                 'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % video_id,
  74                 video_id, 'Downloading video SMIL')
  75
  76             http_base = find_xpath_attr(
  77                 smil,
  78                 './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'),
  79                 'name',
  80                 'httpBase').get('content')
  81
  82             URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=A&g=A'
  83
  84             for video in smil.findall(
  85                     './{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')):
  86                 vbr = int_or_none(video.attrib['system-bitrate'], scale=1000)
  87                 formats.append({
  88                     'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
  89                     'format_id': '%dk' % vbr,
  90                     'vbr': vbr,
  91                 })
  92
  93         self._sort_formats(formats)
  94
  95         return {
  96             'id': '%s' % video_id,
  97             'display_id': display_id,
  98             'title': unescapeHTML(clip_info['clip_title']),
  99             'description': unescapeHTML(clip_info.get('caption')),
 100             'thumbnail': clip_info.get('video_still_url'),
 101             'duration': clip_info.get('duration'),
 102             'formats': formats,
 103         }