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 .subtitles import SubtitlesInfoExtractor
   7 from ..compat import (
   8     compat_str,
   9 )
  10 from ..utils import (
  11     determine_ext,
  12     ExtractorError,
  13     xpath_with_ns,
  14 )
  15
  16 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
  17
  18
  19 class ThePlatformIE(SubtitlesInfoExtractor):
  20     _VALID_URL = r'''(?x)
  21         (?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/
  22            (?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
  23          |theplatform:)(?P<id>[^/\?&]+)'''
  24
  25     _TEST = {
  26         # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
  27         'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
  28         'info_dict': {
  29             'id': 'e9I_cZgTgIPd',
  30             'ext': 'flv',
  31             'title': 'Blackberry\'s big, bold Z30',
  32             'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
  33             'duration': 247,
  34         },
  35         'params': {
  36             # rtmp download
  37             'skip_download': True,
  38         },
  39     }
  40
  41     def _real_extract(self, url):
  42         mobj = re.match(self._VALID_URL, url)
  43         video_id = mobj.group('id')
  44         if mobj.group('config'):
  45             config_url = url + '&form=json'
  46             config_url = config_url.replace('swf/', 'config/')
  47             config_url = config_url.replace('onsite/', 'onsite/config/')
  48             config = self._download_json(config_url, video_id, 'Downloading config')
  49             smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m'
  50         else:
  51             smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
  52                         'format=smil&mbr=true'.format(video_id))
  53
  54         meta = self._download_xml(smil_url, video_id)
  55         try:
  56             error_msg = next(
  57                 n.attrib['abstract']
  58                 for n in meta.findall(_x('.//smil:ref'))
  59                 if n.attrib.get('title') == 'Geographic Restriction')
  60         except StopIteration:
  61             pass
  62         else:
  63             raise ExtractorError(error_msg, expected=True)
  64
  65         info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
  66         info_json = self._download_webpage(info_url, video_id)
  67         info = json.loads(info_json)
  68
  69         subtitles = {}
  70         captions = info.get('captions')
  71         if isinstance(captions, list):
  72             for caption in captions:
  73                 lang, src = caption.get('lang'), caption.get('src')
  74                 if lang and src:
  75                     subtitles[lang] = src
  76
  77         if self._downloader.params.get('listsubtitles', False):
  78             self._list_available_subtitles(video_id, subtitles)
  79             return
  80
  81         subtitles = self.extract_subtitles(video_id, subtitles)
  82
  83         head = meta.find(_x('smil:head'))
  84         body = meta.find(_x('smil:body'))
  85
  86         f4m_node = body.find(_x('smil:seq//smil:video'))
  87         if f4m_node is not None and '.f4m' in f4m_node.attrib['src']:
  88             f4m_url = f4m_node.attrib['src']
  89             if 'manifest.f4m?' not in f4m_url:
  90                 f4m_url += '?'
  91             # the parameters are from syfy.com, other sites may use others,
  92             # they also work for nbc.com
  93             f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3'
  94             formats = self._extract_f4m_formats(f4m_url, video_id)
  95         else:
  96             formats = []
  97             switch = body.find(_x('smil:switch'))
  98             if switch is not None:
  99                 base_url = head.find(_x('smil:meta')).attrib['base']
 100                 for f in switch.findall(_x('smil:video')):
 101                     attr = f.attrib
 102                     width = int(attr['width'])
 103                     height = int(attr['height'])
 104                     vbr = int(attr['system-bitrate']) // 1000
 105                     format_id = '%dx%d_%dk' % (width, height, vbr)
 106                     formats.append({
 107                         'format_id': format_id,
 108                         'url': base_url,
 109                         'play_path': 'mp4:' + attr['src'],
 110                         'ext': 'flv',
 111                         'width': width,
 112                         'height': height,
 113                         'vbr': vbr,
 114                     })
 115             else:
 116                 switch = body.find(_x('smil:seq//smil:switch'))
 117                 for f in switch.findall(_x('smil:video')):
 118                     attr = f.attrib
 119                     vbr = int(attr['system-bitrate']) // 1000
 120                     ext = determine_ext(attr['src'])
 121                     if ext == 'once':
 122                         ext = 'mp4'
 123                     formats.append({
 124                         'format_id': compat_str(vbr),
 125                         'url': attr['src'],
 126                         'vbr': vbr,
 127                         'ext': ext,
 128                     })
 129             self._sort_formats(formats)
 130
 131         return {
 132             'id': video_id,
 133             'title': info['title'],
 134             'subtitles': subtitles,
 135             'formats': formats,
 136             'description': info['description'],
 137             'thumbnail': info['defaultThumbnailUrl'],
 138             'duration': info['duration'] // 1000,
 139         }