youtube_dl/extractor/theplatform.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     xpath_with_ns,
  10 )
  11
  12 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
  13
  14
  15 class ThePlatformIE(InfoExtractor):
  16     _VALID_URL = r'''(?x)
  17         (?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/
  18            (?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
  19          |theplatform:)(?P<id>[^/\?&]+)'''
  20
  21     _TEST = {
  22         # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
  23         'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
  24         'info_dict': {
  25             'id': 'e9I_cZgTgIPd',
  26             'ext': 'flv',
  27             'title': 'Blackberry\'s big, bold Z30',
  28             'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
  29             'duration': 247,
  30         },
  31         'params': {
  32             # rtmp download
  33             'skip_download': True,
  34         },
  35     }
  36
  37     def _get_info(self, video_id, smil_url):
  38         meta = self._download_xml(smil_url, video_id)
  39
  40         try:
  41             error_msg = next(
  42                 n.attrib['abstract']
  43                 for n in meta.findall(_x('.//smil:ref'))
  44                 if n.attrib.get('title') == 'Geographic Restriction')
  45         except StopIteration:
  46             pass
  47         else:
  48             raise ExtractorError(error_msg, expected=True)
  49
  50         info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
  51         info_json = self._download_webpage(info_url, video_id)
  52         info = json.loads(info_json)
  53
  54         head = meta.find(_x('smil:head'))
  55         body = meta.find(_x('smil:body'))
  56
  57         f4m_node = body.find(_x('smil:seq//smil:video'))
  58         if f4m_node is not None:
  59             f4m_url = f4m_node.attrib['src']
  60             if 'manifest.f4m?' not in f4m_url:
  61                 f4m_url += '?'
  62             # the parameters are from syfy.com, other sites may use others,
  63             # they also work for nbc.com
  64             f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3'
  65             formats = [{
  66                 'ext': 'flv',
  67                 'url': f4m_url,
  68             }]
  69         else:
  70             base_url = head.find(_x('smil:meta')).attrib['base']
  71             switch = body.find(_x('smil:switch'))
  72             formats = []
  73             for f in switch.findall(_x('smil:video')):
  74                 attr = f.attrib
  75                 width = int(attr['width'])
  76                 height = int(attr['height'])
  77                 vbr = int(attr['system-bitrate']) // 1000
  78                 format_id = '%dx%d_%dk' % (width, height, vbr)
  79                 formats.append({
  80                     'format_id': format_id,
  81                     'url': base_url,
  82                     'play_path': 'mp4:' + attr['src'],
  83                     'ext': 'flv',
  84                     'width': width,
  85                     'height': height,
  86                     'vbr': vbr,
  87                 })
  88             self._sort_formats(formats)
  89
  90         return {
  91             'id': video_id,
  92             'title': info['title'],
  93             'formats': formats,
  94             'description': info['description'],
  95             'thumbnail': info['defaultThumbnailUrl'],
  96             'duration': info['duration']//1000,
  97         }
  98
  99     def _real_extract(self, url):
 100         mobj = re.match(self._VALID_URL, url)
 101         video_id = mobj.group('id')
 102         if mobj.group('config'):
 103             config_url = url+ '&form=json'
 104             config_url = config_url.replace('swf/', 'config/')
 105             config_url = config_url.replace('onsite/', 'onsite/config/')
 106             config = self._download_json(config_url, video_id, 'Downloading config')
 107             smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m'
 108         else:
 109             smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
 110                 'format=smil&mbr=true'.format(video_id))
 111         return self._get_info(video_id, smil_url)