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