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     determine_ext,
  10 )
  11
  12
  13 class HowStuffWorksIE(InfoExtractor):
  14     _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*(?:\d+-)?(?P<id>.+?)-video\.htm'
  15     _TESTS = [
  16         {
  17             'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm',
  18             'info_dict': {
  19                 'id': '450221',
  20                 'ext': 'flv',
  21                 'title': 'Cool Jobs - Iditarod Musher',
  22                 '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.',
  23                 'display_id': 'cool-jobs-iditarod-musher',
  24                 'thumbnail': r're:^https?://.*\.jpg$',
  25                 'duration': 161,
  26             },
  27             'skip': 'Video broken',
  28         },
  29         {
  30             'url': 'http://adventure.howstuffworks.com/7199-survival-zone-food-and-water-in-the-savanna-video.htm',
  31             'info_dict': {
  32                 'id': '453464',
  33                 'ext': 'mp4',
  34                 'title': 'Survival Zone: Food and Water In the Savanna',
  35                 'description': 'Learn how to find both food and water while trekking in the African savannah. In this video from the Discovery Channel.',
  36                 'display_id': 'survival-zone-food-and-water-in-the-savanna',
  37                 'thumbnail': r're:^https?://.*\.jpg$',
  38             },
  39         },
  40         {
  41             'url': 'http://entertainment.howstuffworks.com/arts/2706-sword-swallowing-1-by-dan-meyer-video.htm',
  42             'info_dict': {
  43                 'id': '440011',
  44                 'ext': 'mp4',
  45                 'title': 'Sword Swallowing #1 by Dan Meyer',
  46                 'description': 'Video footage (1 of 3) used by permission of the owner Dan Meyer through Sword Swallowers Association International <www.swordswallow.org>',
  47                 'display_id': 'sword-swallowing-1-by-dan-meyer',
  48                 'thumbnail': r're:^https?://.*\.jpg$',
  49             },
  50         },
  51         {
  52             'url': 'http://shows.howstuffworks.com/stuff-to-blow-your-mind/optical-illusions-video.htm',
  53             'only_matching': True,
  54         }
  55     ]
  56
  57     def _real_extract(self, url):
  58         display_id = self._match_id(url)
  59         webpage = self._download_webpage(url, display_id)
  60         clip_js = self._search_regex(
  61             r'(?s)var clip = ({.*?});', webpage, 'clip info')
  62         clip_info = self._parse_json(
  63             clip_js, display_id, transform_source=js_to_json)
  64
  65         video_id = clip_info['content_id']
  66         formats = []
  67         m3u8_url = clip_info.get('m3u8')
  68         if m3u8_url and determine_ext(m3u8_url) == 'm3u8':
  69             formats.extend(self._extract_m3u8_formats(m3u8_url, video_id, 'mp4', format_id='hls', fatal=True))
  70         flv_url = clip_info.get('flv_url')
  71         if flv_url:
  72             formats.append({
  73                 'url': flv_url,
  74                 'format_id': 'flv',
  75             })
  76         for video in clip_info.get('mp4', []):
  77             formats.append({
  78                 'url': video['src'],
  79                 'format_id': 'mp4-%s' % video['bitrate'],
  80                 'vbr': int_or_none(video['bitrate'].rstrip('k')),
  81             })
  82
  83         if not formats:
  84             smil = self._download_xml(
  85                 'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % video_id,
  86                 video_id, 'Downloading video SMIL')
  87
  88             http_base = find_xpath_attr(
  89                 smil,
  90                 './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'),
  91                 'name',
  92                 'httpBase').get('content')
  93
  94             URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=A&g=A'
  95
  96             for video in smil.findall(
  97                     './{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')):
  98                 vbr = int_or_none(video.attrib['system-bitrate'], scale=1000)
  99                 formats.append({
 100                     'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
 101                     'format_id': '%dk' % vbr,
 102                     'vbr': vbr,
 103                 })
 104
 105         self._sort_formats(formats)
 106
 107         return {
 108             'id': '%s' % video_id,
 109             'display_id': display_id,
 110             'title': unescapeHTML(clip_info['clip_title']),
 111             'description': unescapeHTML(clip_info.get('caption')),
 112             'thumbnail': clip_info.get('video_still_url'),
 113             'duration': int_or_none(clip_info.get('duration')),
 114             'formats': formats,
 115         }