_ Git - youtube-dl/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_urlencode
  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/v2/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         title = metadata['title']
  26
  27         auth_data = self._download_json(
  28             self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) +
  29             compat_urllib_parse_urlencode({
  30                 'domain': domain,
  31                 'supportedFormats': 'mp4,rtmp,m3u8,hds',
  32             }), video_id)
  33
  34         cur_auth_data = auth_data['authorization_data'][embed_code]
  35
  36         urls = []
  37         formats = []
  38         if cur_auth_data['authorized']:
  39             for stream in cur_auth_data['streams']:
  40                 url = base64.b64decode(
  41                     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                     formats.extend(self._extract_m3u8_formats(
  48                         url, embed_code, 'mp4', 'm3u8_native',
  49                         m3u8_id='hls', fatal=False))
  50                 elif delivery_type == 'hds' or '.f4m' in url:
  51                     formats.extend(self._extract_f4m_formats(
  52                         url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
  53                 elif '.smil' in url:
  54                     formats.extend(self._extract_smil_formats(
  55                         url, embed_code, fatal=False))
  56                 else:
  57                     formats.append({
  58                         'url': url,
  59                         'ext': stream.get('delivery_type'),
  60                         'vcodec': stream.get('video_codec'),
  61                         'format_id': delivery_type,
  62                         'width': int_or_none(stream.get('width')),
  63                         'height': int_or_none(stream.get('height')),
  64                         'abr': int_or_none(stream.get('audio_bitrate')),
  65                         'vbr': int_or_none(stream.get('video_bitrate')),
  66                         'fps': float_or_none(stream.get('framerate')),
  67                     })
  68         else:
  69             raise ExtractorError('%s said: %s' % (
  70                 self.IE_NAME, cur_auth_data['message']), expected=True)
  71         self._sort_formats(formats)
  72
  73         subtitles = {}
  74         for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items():
  75             sub_url = sub.get('url')
  76             if not sub_url:
  77                 continue
  78             subtitles[lang] = [{
  79                 'url': sub_url,
  80             }]
  81
  82         return {
  83             'id': embed_code,
  84             'title': title,
  85             'description': metadata.get('description'),
  86             'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
  87             'duration': float_or_none(metadata.get('duration'), 1000),
  88             'subtitles': subtitles,
  89             'formats': formats,
  90         }
  91
  92
  93 class OoyalaIE(OoyalaBaseIE):
  94     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
  95
  96     _TESTS = [
  97         {
  98             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
  99             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
 100             'info_dict': {
 101                 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
 102                 'ext': 'mp4',
 103                 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
 104                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
 105                 'duration': 853.386,
 106             },
 107             # The video in the original webpage now uses PlayWire
 108             'skip': 'Ooyala said: movie expired',
 109         }, {
 110             # Only available for ipad
 111             'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 112             'info_dict': {
 113                 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 114                 'ext': 'mp4',
 115                 'title': 'Simulation Overview - Levels of Simulation',
 116                 'duration': 194.948,
 117             },
 118         },
 119         {
 120             # Information available only through SAS api
 121             # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
 122             'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 123             'md5': 'a84001441b35ea492bc03736e59e7935',
 124             'info_dict': {
 125                 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 126                 'ext': 'mp4',
 127                 'title': 'Divide Tool Path.mp4',
 128                 'duration': 204.405,
 129             }
 130         }
 131     ]
 132
 133     @staticmethod
 134     def _url_for_embed_code(embed_code):
 135         return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
 136
 137     @classmethod
 138     def _build_url_result(cls, embed_code):
 139         return cls.url_result(cls._url_for_embed_code(embed_code),
 140                               ie=cls.ie_key())
 141
 142     def _real_extract(self, url):
 143         url, smuggled_data = unsmuggle_url(url, {})
 144         embed_code = self._match_id(url)
 145         domain = smuggled_data.get('domain')
 146         content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
 147         return self._extract(content_tree_url, embed_code, domain)
 148
 149
 150 class OoyalaExternalIE(OoyalaBaseIE):
 151     _VALID_URL = r'''(?x)
 152                     (?:
 153                         ooyalaexternal:|
 154                         https?://.+?\.ooyala\.com/.*?\bexternalId=
 155                     )
 156                     (?P<partner_id>[^:]+)
 157                     :
 158                     (?P<id>.+)
 159                     (?:
 160                         :|
 161                         .*?&pcode=
 162                     )
 163                     (?P<pcode>.+?)
 164                     (?:&|$)
 165                     '''
 166
 167     _TEST = {
 168         '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',
 169         'info_dict': {
 170             'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
 171             'ext': 'mp4',
 172             'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
 173             'duration': 1302.0,
 174         },
 175         'params': {
 176             # m3u8 download
 177             'skip_download': True,
 178         },
 179     }
 180
 181     def _real_extract(self, url):
 182         partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
 183         content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
 184         return self._extract(content_tree_url, video_id)