Raphaël G. Git Repositories - youtubedl/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     determine_ext,
  10     ExtractorError,
  11     xpath_with_ns,
  12 )
  13
  14 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
  15
  16
  17 class ThePlatformIE(InfoExtractor):
  18     _VALID_URL = r'''(?x)
  19         (?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/
  20            (?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
  21          |theplatform:)(?P<id>[^/\?&]+)'''
  22
  23     _TEST = {
  24         # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
  25         'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
  26         'info_dict': {
  27             'id': 'e9I_cZgTgIPd',
  28             'ext': 'flv',
  29             'title': 'Blackberry\'s big, bold Z30',
  30             'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
  31             'duration': 247,
  32         },
  33         'params': {
  34             # rtmp download
  35             'skip_download': True,
  36         },
  37     }
  38     def _real_extract(self, url):
  39         mobj = re.match(self._VALID_URL, url)
  40         video_id = mobj.group('id')
  41         if mobj.group('config'):
  42             config_url = url+ '&form=json'
  43             config_url = config_url.replace('swf/', 'config/')
  44             config_url = config_url.replace('onsite/', 'onsite/config/')
  45             config = self._download_json(config_url, video_id, 'Downloading config')
  46             smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m'
  47         else:
  48             smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
  49                 'format=smil&mbr=true'.format(video_id))
  50
  51
  52         meta = self._download_xml(smil_url, video_id)
  53         try:
  54             error_msg = next(
  55                 n.attrib['abstract']
  56                 for n in meta.findall(_x('.//smil:ref'))
  57                 if n.attrib.get('title') == 'Geographic Restriction')
  58         except StopIteration:
  59             pass
  60         else:
  61             raise ExtractorError(error_msg, expected=True)
  62
  63         info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
  64         info_json = self._download_webpage(info_url, video_id)
  65         info = json.loads(info_json)
  66
  67         head = meta.find(_x('smil:head'))
  68         body = meta.find(_x('smil:body'))
  69
  70         f4m_node = body.find(_x('smil:seq//smil:video'))
  71         if f4m_node is not None and '.f4m' in f4m_node.attrib['src']:
  72             f4m_url = f4m_node.attrib['src']
  73             if 'manifest.f4m?' not in f4m_url:
  74                 f4m_url += '?'
  75             # the parameters are from syfy.com, other sites may use others,
  76             # they also work for nbc.com
  77             f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3'
  78             formats = self._extract_f4m_formats(f4m_url, video_id)
  79         else:
  80             formats = []
  81             switch = body.find(_x('smil:switch'))
  82             if switch is not None:
  83                 base_url = head.find(_x('smil:meta')).attrib['base']
  84                 for f in switch.findall(_x('smil:video')):
  85                     attr = f.attrib
  86                     width = int(attr['width'])
  87                     height = int(attr['height'])
  88                     vbr = int(attr['system-bitrate']) // 1000
  89                     format_id = '%dx%d_%dk' % (width, height, vbr)
  90                     formats.append({
  91                         'format_id': format_id,
  92                         'url': base_url,
  93                         'play_path': 'mp4:' + attr['src'],
  94                         'ext': 'flv',
  95                         'width': width,
  96                         'height': height,
  97                         'vbr': vbr,
  98                     })
  99             else:
 100                 switch = body.find(_x('smil:seq//smil:switch'))
 101                 for f in switch.findall(_x('smil:video')):
 102                     attr = f.attrib
 103                     vbr = int(attr['system-bitrate']) // 1000
 104                     ext = determine_ext(attr['src'])
 105                     if ext == 'once':
 106                         ext = 'mp4'
 107                     formats.append({
 108                         'format_id': compat_str(vbr),
 109                         'url': attr['src'],
 110                         'vbr': vbr,
 111                         'ext': ext,
 112                     })
 113             self._sort_formats(formats)
 114
 115         return {
 116             'id': video_id,
 117             'title': info['title'],
 118             'formats': formats,
 119             'description': info['description'],
 120             'thumbnail': info['defaultThumbnailUrl'],
 121             'duration': info['duration']//1000,
 122         }