youtube_dl/extractor/ooyala.py

   1 from __future__ import unicode_literals
   2 import re
   3 import base64
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     int_or_none,
   8     float_or_none,
   9     ExtractorError,
  10     unsmuggle_url,
  11 )
  12 from ..compat import compat_urllib_parse
  13
  14
  15 class OoyalaBaseIE(InfoExtractor):
  16
  17     def _extract(self, content_tree_url, video_id, domain='example.org'):
  18         content_tree = self._download_json(content_tree_url, video_id)['content_tree']
  19         metadata = content_tree[list(content_tree)[0]]
  20         embed_code = metadata['embed_code']
  21         pcode = metadata.get('asset_pcode') or embed_code
  22         video_info = {
  23             'id': embed_code,
  24             'title': metadata['title'],
  25             'description': metadata.get('description'),
  26             'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
  27             'duration': float_or_none(metadata.get('duration'), 1000),
  28         }
  29
  30         urls = []
  31         formats = []
  32         for supported_format in ('mp4', 'm3u8', 'hds', 'rtmp'):
  33             auth_data = self._download_json(
  34                 'http://player.ooyala.com/sas/player_api/v1/authorization/embed_code/%s/%s?' % (pcode, embed_code) + compat_urllib_parse.urlencode({'domain': domain, 'supportedFormats': supported_format}),
  35                 video_id, 'Downloading %s JSON' % supported_format)
  36
  37             cur_auth_data = auth_data['authorization_data'][embed_code]
  38
  39             if cur_auth_data['authorized']:
  40                 for stream in cur_auth_data['streams']:
  41                     url = base64.b64decode(stream['url']['data'].encode('ascii')).decode('utf-8')
  42                     if url in urls:
  43                         continue
  44                     urls.append(url)
  45                     delivery_type = stream['delivery_type']
  46                     if delivery_type == 'hls' or '.m3u8' in url:
  47                         m3u8_formats = self._extract_m3u8_formats(url, embed_code, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
  48                         if m3u8_formats:
  49                             formats.extend(m3u8_formats)
  50                     elif delivery_type == 'hds' or '.f4m' in url:
  51                         f4m_formats = self._extract_f4m_formats(url, embed_code, f4m_id='hds', fatal=False)
  52                         if f4m_formats:
  53                             formats.extend(f4m_formats)
  54                     elif '.smil' in url:
  55                         smil_formats = self._extract_smil_formats(url, embed_code, fatal=False)
  56                         if smil_formats:
  57                             formats.extend(smil_formats)
  58                     else:
  59                         formats.append({
  60                             'url': url,
  61                             'ext': stream.get('delivery_type'),
  62                             'vcodec': stream.get('video_codec'),
  63                             'format_id': delivery_type,
  64                             'width': int_or_none(stream.get('width')),
  65                             'height': int_or_none(stream.get('height')),
  66                             'abr': int_or_none(stream.get('audio_bitrate')),
  67                             'vbr': int_or_none(stream.get('video_bitrate')),
  68                             'fps': float_or_none(stream.get('framerate')),
  69                         })
  70             else:
  71                 raise ExtractorError('%s said: %s' % (self.IE_NAME, cur_auth_data['message']), expected=True)
  72         self._sort_formats(formats)
  73
  74         video_info['formats'] = formats
  75         return video_info
  76
  77
  78 class OoyalaIE(OoyalaBaseIE):
  79     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
  80
  81     _TESTS = [
  82         {
  83             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
  84             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  85             'info_dict': {
  86                 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  87                 'ext': 'mp4',
  88                 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
  89                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
  90                 'duration': 853.386,
  91             },
  92         }, {
  93             # Only available for ipad
  94             'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  95             'info_dict': {
  96                 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  97                 'ext': 'mp4',
  98                 'title': 'Simulation Overview - Levels of Simulation',
  99                 'duration': 194.948,
 100             },
 101         },
 102         {
 103             # Information available only through SAS api
 104             # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
 105             'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 106             'md5': 'a84001441b35ea492bc03736e59e7935',
 107             'info_dict': {
 108                 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 109                 'ext': 'mp4',
 110                 'title': 'Divide Tool Path.mp4',
 111                 'duration': 204.405,
 112             }
 113         }
 114     ]
 115
 116     @staticmethod
 117     def _url_for_embed_code(embed_code):
 118         return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
 119
 120     @classmethod
 121     def _build_url_result(cls, embed_code):
 122         return cls.url_result(cls._url_for_embed_code(embed_code),
 123                               ie=cls.ie_key())
 124
 125     def _real_extract(self, url):
 126         url, smuggled_data = unsmuggle_url(url, {})
 127         embed_code = self._match_id(url)
 128         domain = smuggled_data.get('domain')
 129         content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/embed_code/%s/%s' % (embed_code, embed_code)
 130         return self._extract(content_tree_url, embed_code, domain)
 131
 132
 133 class OoyalaExternalIE(OoyalaBaseIE):
 134     _VALID_URL = r'''(?x)
 135                     (?:
 136                         ooyalaexternal:|
 137                         https?://.+?\.ooyala\.com/.*?\bexternalId=
 138                     )
 139                     (?P<partner_id>[^:]+)
 140                     :
 141                     (?P<id>.+)
 142                     (?:
 143                         :|
 144                         .*?&pcode=
 145                     )
 146                     (?P<pcode>.+?)
 147                     (?:&|$)
 148                     '''
 149
 150     _TEST = {
 151         'url': 'https://player.ooyala.com/player.js?externalId=espn:10365079&pcode=1kNG061cgaoolOncv54OAO1ceO-I&adSetCode=91cDU6NuXTGKz3OdjOxFdAgJVtQcKJnI&callback=handleEvents&hasModuleParams=1&height=968&playerBrandingId=7af3bd04449c444c964f347f11873075&targetReplaceId=videoPlayer&width=1656&wmode=opaque&allowScriptAccess=always',
 152         'info_dict': {
 153             'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
 154             'ext': 'mp4',
 155             'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
 156             'duration': 1302000,
 157         },
 158         'params': {
 159             # m3u8 download
 160             'skip_download': True,
 161         },
 162     }
 163
 164     def _real_extract(self, url):
 165         partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
 166         content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/external_id/%s/%s:%s' % (pcode, partner_id, video_id)
 167         return self._extract(content_tree_url, video_id)