Raphaël G. Git Repositories - youtubedl/blob - 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     _PLAYER_BASE = 'http://player.ooyala.com/'
  17     _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
  18     _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v1/authorization/embed_code/%s/%s?'
  19
  20     def _extract(self, content_tree_url, video_id, domain='example.org'):
  21         content_tree = self._download_json(content_tree_url, video_id)['content_tree']
  22         metadata = content_tree[list(content_tree)[0]]
  23         embed_code = metadata['embed_code']
  24         pcode = metadata.get('asset_pcode') or embed_code
  25         video_info = {
  26             'id': embed_code,
  27             'title': metadata['title'],
  28             'description': metadata.get('description'),
  29             'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
  30             'duration': float_or_none(metadata.get('duration'), 1000),
  31         }
  32
  33         urls = []
  34         formats = []
  35         for supported_format in ('mp4', 'm3u8', 'hds', 'rtmp'):
  36             auth_data = self._download_json(
  37                 self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) +
  38                 compat_urllib_parse.urlencode({
  39                     'domain': domain,
  40                     'supportedFormats': supported_format
  41                 }),
  42                 video_id, 'Downloading %s JSON' % supported_format)
  43
  44             cur_auth_data = auth_data['authorization_data'][embed_code]
  45
  46             if cur_auth_data['authorized']:
  47                 for stream in cur_auth_data['streams']:
  48                     url = base64.b64decode(
  49                         stream['url']['data'].encode('ascii')).decode('utf-8')
  50                     if url in urls:
  51                         continue
  52                     urls.append(url)
  53                     delivery_type = stream['delivery_type']
  54                     if delivery_type == 'hls' or '.m3u8' in url:
  55                         formats.extend(self._extract_m3u8_formats(
  56                             url, embed_code, 'mp4', 'm3u8_native',
  57                             m3u8_id='hls', fatal=False))
  58                     elif delivery_type == 'hds' or '.f4m' in url:
  59                         formats.extend(self._extract_f4m_formats(
  60                             url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
  61                     elif '.smil' in url:
  62                         formats.extend(self._extract_smil_formats(
  63                             url, embed_code, fatal=False))
  64                     else:
  65                         formats.append({
  66                             'url': url,
  67                             'ext': stream.get('delivery_type'),
  68                             'vcodec': stream.get('video_codec'),
  69                             'format_id': delivery_type,
  70                             'width': int_or_none(stream.get('width')),
  71                             'height': int_or_none(stream.get('height')),
  72                             'abr': int_or_none(stream.get('audio_bitrate')),
  73                             'vbr': int_or_none(stream.get('video_bitrate')),
  74                             'fps': float_or_none(stream.get('framerate')),
  75                         })
  76             else:
  77                 raise ExtractorError('%s said: %s' % (
  78                     self.IE_NAME, cur_auth_data['message']), expected=True)
  79         self._sort_formats(formats)
  80
  81         video_info['formats'] = formats
  82         return video_info
  83
  84
  85 class OoyalaIE(OoyalaBaseIE):
  86     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
  87
  88     _TESTS = [
  89         {
  90             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
  91             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  92             'info_dict': {
  93                 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  94                 'ext': 'mp4',
  95                 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
  96                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
  97                 'duration': 853.386,
  98             },
  99         }, {
 100             # Only available for ipad
 101             'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 102             'info_dict': {
 103                 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 104                 'ext': 'mp4',
 105                 'title': 'Simulation Overview - Levels of Simulation',
 106                 'duration': 194.948,
 107             },
 108         },
 109         {
 110             # Information available only through SAS api
 111             # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
 112             'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 113             'md5': 'a84001441b35ea492bc03736e59e7935',
 114             'info_dict': {
 115                 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 116                 'ext': 'mp4',
 117                 'title': 'Divide Tool Path.mp4',
 118                 'duration': 204.405,
 119             }
 120         }
 121     ]
 122
 123     @staticmethod
 124     def _url_for_embed_code(embed_code):
 125         return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
 126
 127     @classmethod
 128     def _build_url_result(cls, embed_code):
 129         return cls.url_result(cls._url_for_embed_code(embed_code),
 130                               ie=cls.ie_key())
 131
 132     def _real_extract(self, url):
 133         url, smuggled_data = unsmuggle_url(url, {})
 134         embed_code = self._match_id(url)
 135         domain = smuggled_data.get('domain')
 136         content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
 137         return self._extract(content_tree_url, embed_code, domain)
 138
 139
 140 class OoyalaExternalIE(OoyalaBaseIE):
 141     _VALID_URL = r'''(?x)
 142                     (?:
 143                         ooyalaexternal:|
 144                         https?://.+?\.ooyala\.com/.*?\bexternalId=
 145                     )
 146                     (?P<partner_id>[^:]+)
 147                     :
 148                     (?P<id>.+)
 149                     (?:
 150                         :|
 151                         .*?&pcode=
 152                     )
 153                     (?P<pcode>.+?)
 154                     (?:&|$)
 155                     '''
 156
 157     _TEST = {
 158         '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',
 159         'info_dict': {
 160             'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
 161             'ext': 'mp4',
 162             'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
 163             'duration': 1302.0,
 164         },
 165         'params': {
 166             # m3u8 download
 167             'skip_download': True,
 168         },
 169     }
 170
 171     def _real_extract(self, url):
 172         partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
 173         content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
 174         return self._extract(content_tree_url, video_id)