youtube_dl/extractor/theplatform.py

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