youtube_dl/extractor/howstuffworks.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5 import random
   6 import string
   7
   8 from .common import InfoExtractor
   9 from ..utils import find_xpath_attr
  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                 'display_id': 'cool-jobs-iditarod-musher',
  20                 'ext': 'flv',
  21                 'title': 'Cool Jobs - Iditarod Musher',
  22                 'description': 'md5:82bb58438a88027b8186a1fccb365f90',
  23                 'thumbnail': 're:^https?://.*\.jpg$',
  24             },
  25             'params': {
  26                 # md5 is not consistent
  27                 'skip_download': True
  28             }
  29         },
  30         {
  31             'url': 'http://adventure.howstuffworks.com/39516-deadliest-catch-jakes-farewell-pots-video.htm',
  32             'info_dict': {
  33                 'id': '553470',
  34                 'display_id': 'deadliest-catch-jakes-farewell-pots',
  35                 'ext': 'mp4',
  36                 'title': 'Deadliest Catch: Jake\'s Farewell Pots',
  37                 'description': 'md5:9632c346d5e43ee238028c9cefd8dbbc',
  38                 'thumbnail': 're:^https?://.*\.jpg$',
  39             },
  40             'params': {
  41                 # md5 is not consistent
  42                 'skip_download': True
  43             }
  44         },
  45         {
  46             'url': 'http://entertainment.howstuffworks.com/arts/2706-sword-swallowing-1-by-dan-meyer-video.htm',
  47             'info_dict': {
  48                 'id': '440011',
  49                 'display_id': 'sword-swallowing-1-by-dan-meyer',
  50                 'ext': 'flv',
  51                 'title': 'Sword Swallowing #1 by Dan Meyer',
  52                 'description': 'md5:b2409e88172913e2e7d3d1159b0ef735',
  53                 'thumbnail': 're:^https?://.*\.jpg$',
  54             },
  55             'params': {
  56                 # md5 is not consistent
  57                 'skip_download': True
  58             }
  59         },
  60     ]
  61
  62     def _real_extract(self, url):
  63         mobj = re.match(self._VALID_URL, url)
  64         display_id = mobj.group('id')
  65         webpage = self._download_webpage(url, display_id)
  66
  67         content_id = self._search_regex(r'var siteSectionId="(\d+)";', webpage, 'content id')
  68
  69         mp4 = self._search_regex(
  70             r'''(?xs)var\s+clip\s*=\s*{\s*
  71                 .+?\s*
  72                 content_id\s*:\s*%s\s*,\s*
  73                 .+?\s*
  74                 mp4\s*:\s*\[(.*?),?\]\s*
  75                 };\s*
  76                 videoData\.push\(clip\);''' % content_id,
  77             webpage, 'mp4', fatal=False, default=None)
  78
  79         smil = self._download_xml(
  80             'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % content_id,
  81             content_id, 'Downloading video SMIL')
  82
  83         http_base = find_xpath_attr(
  84             smil,
  85             './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'),
  86             'name',
  87             'httpBase').get('content')
  88
  89         def random_string(str_len=0):
  90             return ''.join([random.choice(string.ascii_uppercase) for _ in range(str_len)])
  91
  92         URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=%s&g=%s' % (random_string(5), random_string(12))
  93
  94         formats = []
  95
  96         if mp4:
  97             for video in json.loads('[%s]' % mp4):
  98                 bitrate = video['bitrate']
  99                 fmt = {
 100                     'url': video['src'].replace('http://pmd.video.howstuffworks.com', http_base) + URL_SUFFIX,
 101                     'format_id': bitrate,
 102                 }
 103                 m = re.search(r'(?P<vbr>\d+)[Kk]', bitrate)
 104                 if m:
 105                     fmt['vbr'] = int(m.group('vbr'))
 106                 formats.append(fmt)
 107         else:
 108             for video in smil.findall(
 109                     './/{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')):
 110                 vbr = int(video.attrib['system-bitrate']) / 1000
 111                 formats.append({
 112                     'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
 113                     'format_id': '%dk' % vbr,
 114                     'vbr': vbr,
 115                 })
 116
 117         self._sort_formats(formats)
 118
 119         title = self._og_search_title(webpage)
 120         TITLE_SUFFIX = ' : HowStuffWorks'
 121         if title.endswith(TITLE_SUFFIX):
 122             title = title[:-len(TITLE_SUFFIX)]
 123
 124         description = self._og_search_description(webpage)
 125         thumbnail = self._og_search_thumbnail(webpage)
 126
 127         return {
 128             'id': content_id,
 129             'display_id': display_id,
 130             'title': title,
 131             'description': description,
 132             'thumbnail': thumbnail,
 133             'formats': formats,
 134         }