_ Git - youtube-dl/blob - youtube_dl/extractor/theplatform.py

   1 import re
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     xpath_with_ns,
   7     find_xpath_attr,
   8 )
   9
  10 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
  11
  12
  13 class ThePlatformIE(InfoExtractor):
  14     _VALID_URL = r'(?:https?://link\.theplatform\.com/s/[^/]+/|theplatform:)(?P<id>[^/\?]+)'
  15
  16     _TEST = {
  17         # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
  18         u'url': u'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
  19         u'info_dict': {
  20             u'id': u'e9I_cZgTgIPd',
  21             u'ext': u'flv',
  22             u'title': u'Blackberry\'s big, bold Z30',
  23             u'description': u'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
  24             u'duration': 247,
  25         },
  26         u'params': {
  27             # rtmp download
  28             u'skip_download': True,
  29         },
  30     }
  31
  32     def _get_info(self, video_id):
  33         smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
  34             'format=smil&mbr=true'.format(video_id))
  35         meta = self._download_xml(smil_url, video_id)
  36         info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
  37         info_json = self._download_webpage(info_url, video_id)
  38         info = json.loads(info_json)
  39
  40         head = meta.find(_x('smil:head'))
  41         body = meta.find(_x('smil:body'))
  42         base_url = head.find(_x('smil:meta')).attrib['base']
  43         switch = body.find(_x('smil:switch'))
  44         formats = []
  45         for f in switch.findall(_x('smil:video')):
  46             attr = f.attrib
  47             formats.append({
  48                 'url': base_url,
  49                 'play_path': 'mp4:' + attr['src'],
  50                 'ext': 'flv',
  51                 'width': int(attr['width']),
  52                 'height': int(attr['height']),
  53                 'vbr': int(attr['system-bitrate']),
  54             })
  55         formats.sort(key=lambda f: (f['height'], f['width'], f['vbr']))
  56
  57         return {
  58             'id': video_id,
  59             'title': info['title'],
  60             'formats': formats,
  61             'description': info['description'],
  62             'thumbnail': info['defaultThumbnailUrl'],
  63             'duration': info['duration']//1000,
  64         }
  65
  66     def _real_extract(self, url):
  67         mobj = re.match(self._VALID_URL, url)
  68         video_id = mobj.group('id')
  69         return self._get_info(video_id)