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     determine_ext,
  12 )
  13 from ..compat import compat_urllib_parse_urlencode
  14
  15
  16 class OoyalaBaseIE(InfoExtractor):
  17     _PLAYER_BASE = 'http://player.ooyala.com/'
  18     _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
  19     _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s?'
  20
  21     def _extract(self, content_tree_url, video_id, domain='example.org'):
  22         content_tree = self._download_json(content_tree_url, video_id)['content_tree']
  23         metadata = content_tree[list(content_tree)[0]]
  24         embed_code = metadata['embed_code']
  25         pcode = metadata.get('asset_pcode') or embed_code
  26         title = metadata['title']
  27
  28         auth_data = self._download_json(
  29             self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) +
  30             compat_urllib_parse_urlencode({
  31                 'domain': domain,
  32                 'supportedFormats': 'mp4,rtmp,m3u8,hds',
  33             }), video_id)
  34
  35         cur_auth_data = auth_data['authorization_data'][embed_code]
  36
  37         urls = []
  38         formats = []
  39         if cur_auth_data['authorized']:
  40             for stream in cur_auth_data['streams']:
  41                 s_url = base64.b64decode(
  42                     stream['url']['data'].encode('ascii')).decode('utf-8')
  43                 if s_url in urls:
  44                     continue
  45                 urls.append(s_url)
  46                 ext = determine_ext(s_url, None)
  47                 delivery_type = stream['delivery_type']
  48                 if delivery_type == 'hls' or ext == 'm3u8':
  49                     formats.extend(self._extract_m3u8_formats(
  50                         s_url, embed_code, 'mp4', 'm3u8_native',
  51                         m3u8_id='hls', fatal=False))
  52                 elif delivery_type == 'hds' or ext == 'f4m':
  53                     formats.extend(self._extract_f4m_formats(
  54                         s_url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
  55                 elif ext == 'smil':
  56                     formats.extend(self._extract_smil_formats(
  57                         s_url, embed_code, fatal=False))
  58                 else:
  59                     formats.append({
  60                         'url': s_url,
  61                         'ext': ext or 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' % (
  72                 self.IE_NAME, cur_auth_data['message']), expected=True)
  73         self._sort_formats(formats)
  74
  75         subtitles = {}
  76         for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items():
  77             sub_url = sub.get('url')
  78             if not sub_url:
  79                 continue
  80             subtitles[lang] = [{
  81                 'url': sub_url,
  82             }]
  83
  84         return {
  85             'id': embed_code,
  86             'title': title,
  87             'description': metadata.get('description'),
  88             'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
  89             'duration': float_or_none(metadata.get('duration'), 1000),
  90             'subtitles': subtitles,
  91             'formats': formats,
  92         }
  93
  94
  95 class OoyalaIE(OoyalaBaseIE):
  96     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
  97
  98     _TESTS = [
  99         {
 100             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
 101             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
 102             'info_dict': {
 103                 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
 104                 'ext': 'mp4',
 105                 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
 106                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
 107                 'duration': 853.386,
 108             },
 109             # The video in the original webpage now uses PlayWire
 110             'skip': 'Ooyala said: movie expired',
 111         }, {
 112             # Only available for ipad
 113             'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 114             'info_dict': {
 115                 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 116                 'ext': 'mp4',
 117                 'title': 'Simulation Overview - Levels of Simulation',
 118                 'duration': 194.948,
 119             },
 120         },
 121         {
 122             # Information available only through SAS api
 123             # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
 124             'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 125             'md5': 'a84001441b35ea492bc03736e59e7935',
 126             'info_dict': {
 127                 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 128                 'ext': 'mp4',
 129                 'title': 'Divide Tool Path.mp4',
 130                 'duration': 204.405,
 131             }
 132         }
 133     ]
 134
 135     @staticmethod
 136     def _url_for_embed_code(embed_code):
 137         return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
 138
 139     @classmethod
 140     def _build_url_result(cls, embed_code):
 141         return cls.url_result(cls._url_for_embed_code(embed_code),
 142                               ie=cls.ie_key())
 143
 144     def _real_extract(self, url):
 145         url, smuggled_data = unsmuggle_url(url, {})
 146         embed_code = self._match_id(url)
 147         domain = smuggled_data.get('domain')
 148         content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
 149         return self._extract(content_tree_url, embed_code, domain)
 150
 151
 152 class OoyalaExternalIE(OoyalaBaseIE):
 153     _VALID_URL = r'''(?x)
 154                     (?:
 155                         ooyalaexternal:|
 156                         https?://.+?\.ooyala\.com/.*?\bexternalId=
 157                     )
 158                     (?P<partner_id>[^:]+)
 159                     :
 160                     (?P<id>.+)
 161                     (?:
 162                         :|
 163                         .*?&pcode=
 164                     )
 165                     (?P<pcode>.+?)
 166                     (?:&|$)
 167                     '''
 168
 169     _TEST = {
 170         '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',
 171         'info_dict': {
 172             'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
 173             'ext': 'mp4',
 174             'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
 175             'duration': 1302.0,
 176         },
 177         'params': {
 178             # m3u8 download
 179             'skip_download': True,
 180         },
 181     }
 182
 183     def _real_extract(self, url):
 184         partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
 185         content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
 186         return self._extract(content_tree_url, video_id)