_ Git - youtube-dl/blob - 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     compat_str,
   9     ExtractorError,
  10     xpath_with_ns,
  11 )
  12
  13 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
  14
  15
  16 class ThePlatformIE(InfoExtractor):
  17     _VALID_URL = r'''(?x)
  18         (?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/
  19            (?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
  20          |theplatform:)(?P<id>[^/\?&]+)'''
  21
  22     _TEST = {
  23         # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
  24         'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
  25         'info_dict': {
  26             'id': 'e9I_cZgTgIPd',
  27             'ext': 'flv',
  28             'title': 'Blackberry\'s big, bold Z30',
  29             'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
  30             'duration': 247,
  31         },
  32         'params': {
  33             # rtmp download
  34             'skip_download': True,
  35         },
  36     }
  37
  38     def _get_info(self, video_id, smil_url):
  39         meta = self._download_xml(smil_url, video_id)
  40
  41         try:
  42             error_msg = next(
  43                 n.attrib['abstract']
  44                 for n in meta.findall(_x('.//smil:ref'))
  45                 if n.attrib.get('title') == 'Geographic Restriction')
  46         except StopIteration:
  47             pass
  48         else:
  49             raise ExtractorError(error_msg, expected=True)
  50
  51         info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
  52         info_json = self._download_webpage(info_url, video_id)
  53         info = json.loads(info_json)
  54
  55         head = meta.find(_x('smil:head'))
  56         body = meta.find(_x('smil:body'))
  57
  58         f4m_node = body.find(_x('smil:seq//smil:video'))
  59         if f4m_node is not None and '.f4m' in f4m_node.attrib['src']:
  60             f4m_url = f4m_node.attrib['src']
  61             if 'manifest.f4m?' not in f4m_url:
  62                 f4m_url += '?'
  63             # the parameters are from syfy.com, other sites may use others,
  64             # they also work for nbc.com
  65             f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3'
  66             formats = self._extract_f4m_formats(f4m_url, video_id)
  67         else:
  68             formats = []
  69             switch = body.find(_x('smil:switch'))
  70             if switch is not None:
  71                 base_url = head.find(_x('smil:meta')).attrib['base']
  72                 for f in switch.findall(_x('smil:video')):
  73                     attr = f.attrib
  74                     width = int(attr['width'])
  75                     height = int(attr['height'])
  76                     vbr = int(attr['system-bitrate']) // 1000
  77                     format_id = '%dx%d_%dk' % (width, height, vbr)
  78                     formats.append({
  79                         'format_id': format_id,
  80                         'url': base_url,
  81                         'play_path': 'mp4:' + attr['src'],
  82                         'ext': 'flv',
  83                         'width': width,
  84                         'height': height,
  85                         'vbr': vbr,
  86                     })
  87             else:
  88                 switch = body.find(_x('smil:seq//smil:switch'))
  89                 for f in switch.findall(_x('smil:video')):
  90                     attr = f.attrib
  91                     vbr = int(attr['system-bitrate']) // 1000
  92                     formats.append({
  93                         'format_id': compat_str(vbr),
  94                         'url': attr['src'],
  95                         'vbr': vbr,
  96                     })
  97             self._sort_formats(formats)
  98
  99         return {
 100             'id': video_id,
 101             'title': info['title'],
 102             'formats': formats,
 103             'description': info['description'],
 104             'thumbnail': info['defaultThumbnailUrl'],
 105             'duration': info['duration']//1000,
 106         }
 107
 108     def _real_extract(self, url):
 109         mobj = re.match(self._VALID_URL, url)
 110         video_id = mobj.group('id')
 111         if mobj.group('config'):
 112             config_url = url+ '&form=json'
 113             config_url = config_url.replace('swf/', 'config/')
 114             config_url = config_url.replace('onsite/', 'onsite/config/')
 115             config = self._download_json(config_url, video_id, 'Downloading config')
 116             smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m'
 117         else:
 118             smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
 119                 'format=smil&mbr=true'.format(video_id))
 120         return self._get_info(video_id, smil_url)