_ 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 ..compat import compat_str
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10     float_or_none,
  11     int_or_none,
  12     try_get,
  13     unsmuggle_url,
  14 )
  15 from ..compat import compat_urllib_parse_urlencode
  16
  17
  18 class OoyalaBaseIE(InfoExtractor):
  19     _PLAYER_BASE = 'http://player.ooyala.com/'
  20     _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
  21     _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s?'
  22
  23     def _extract(self, content_tree_url, video_id, domain='example.org', supportedformats=None, embed_token=None):
  24         content_tree = self._download_json(content_tree_url, video_id)['content_tree']
  25         metadata = content_tree[list(content_tree)[0]]
  26         embed_code = metadata['embed_code']
  27         pcode = metadata.get('asset_pcode') or embed_code
  28         title = metadata['title']
  29
  30         auth_data = self._download_json(
  31             self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) +
  32             compat_urllib_parse_urlencode({
  33                 'domain': domain,
  34                 'supportedFormats': supportedformats or 'mp4,rtmp,m3u8,hds,dash,smooth',
  35                 'embedToken': embed_token,
  36             }), video_id)
  37
  38         cur_auth_data = auth_data['authorization_data'][embed_code]
  39
  40         urls = []
  41         formats = []
  42         if cur_auth_data['authorized']:
  43             for stream in cur_auth_data['streams']:
  44                 url_data = try_get(stream, lambda x: x['url']['data'], compat_str)
  45                 if not url_data:
  46                     continue
  47                 s_url = base64.b64decode(url_data.encode('ascii')).decode('utf-8')
  48                 if not s_url or s_url in urls:
  49                     continue
  50                 urls.append(s_url)
  51                 ext = determine_ext(s_url, None)
  52                 delivery_type = stream.get('delivery_type')
  53                 if delivery_type == 'hls' or ext == 'm3u8':
  54                     formats.extend(self._extract_m3u8_formats(
  55                         re.sub(r'/ip(?:ad|hone)/', '/all/', s_url), embed_code, 'mp4', 'm3u8_native',
  56                         m3u8_id='hls', fatal=False))
  57                 elif delivery_type == 'hds' or ext == 'f4m':
  58                     formats.extend(self._extract_f4m_formats(
  59                         s_url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
  60                 elif delivery_type == 'dash' or ext == 'mpd':
  61                     formats.extend(self._extract_mpd_formats(
  62                         s_url, embed_code, mpd_id='dash', fatal=False))
  63                 elif delivery_type == 'smooth':
  64                     self._extract_ism_formats(
  65                         s_url, embed_code, ism_id='mss', fatal=False)
  66                 elif ext == 'smil':
  67                     formats.extend(self._extract_smil_formats(
  68                         s_url, embed_code, fatal=False))
  69                 else:
  70                     formats.append({
  71                         'url': s_url,
  72                         'ext': ext or delivery_type,
  73                         'vcodec': stream.get('video_codec'),
  74                         'format_id': delivery_type,
  75                         'width': int_or_none(stream.get('width')),
  76                         'height': int_or_none(stream.get('height')),
  77                         'abr': int_or_none(stream.get('audio_bitrate')),
  78                         'vbr': int_or_none(stream.get('video_bitrate')),
  79                         'fps': float_or_none(stream.get('framerate')),
  80                     })
  81         else:
  82             raise ExtractorError('%s said: %s' % (
  83                 self.IE_NAME, cur_auth_data['message']), expected=True)
  84         self._sort_formats(formats)
  85
  86         subtitles = {}
  87         for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items():
  88             sub_url = sub.get('url')
  89             if not sub_url:
  90                 continue
  91             subtitles[lang] = [{
  92                 'url': sub_url,
  93             }]
  94
  95         return {
  96             'id': embed_code,
  97             'title': title,
  98             'description': metadata.get('description'),
  99             'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
 100             'duration': float_or_none(metadata.get('duration'), 1000),
 101             'subtitles': subtitles,
 102             'formats': formats,
 103         }
 104
 105
 106 class OoyalaIE(OoyalaBaseIE):
 107     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
 108
 109     _TESTS = [
 110         {
 111             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
 112             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
 113             'info_dict': {
 114                 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
 115                 'ext': 'mp4',
 116                 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
 117                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
 118                 'duration': 853.386,
 119             },
 120             # The video in the original webpage now uses PlayWire
 121             'skip': 'Ooyala said: movie expired',
 122         }, {
 123             # Only available for ipad
 124             'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 125             'info_dict': {
 126                 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 127                 'ext': 'mp4',
 128                 'title': 'Simulation Overview - Levels of Simulation',
 129                 'duration': 194.948,
 130             },
 131         },
 132         {
 133             # Information available only through SAS api
 134             # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
 135             'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 136             'md5': 'a84001441b35ea492bc03736e59e7935',
 137             'info_dict': {
 138                 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 139                 'ext': 'mp4',
 140                 'title': 'Divide Tool Path.mp4',
 141                 'duration': 204.405,
 142             }
 143         },
 144         {
 145             # empty stream['url']['data']
 146             'url': 'http://player.ooyala.com/player.js?embedCode=w2bnZtYjE6axZ_dw1Cd0hQtXd_ige2Is',
 147             'only_matching': True,
 148         }
 149     ]
 150
 151     @staticmethod
 152     def _url_for_embed_code(embed_code):
 153         return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
 154
 155     @classmethod
 156     def _build_url_result(cls, embed_code):
 157         return cls.url_result(cls._url_for_embed_code(embed_code),
 158                               ie=cls.ie_key())
 159
 160     def _real_extract(self, url):
 161         url, smuggled_data = unsmuggle_url(url, {})
 162         embed_code = self._match_id(url)
 163         domain = smuggled_data.get('domain')
 164         supportedformats = smuggled_data.get('supportedformats')
 165         embed_token = smuggled_data.get('embed_token')
 166         content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
 167         return self._extract(content_tree_url, embed_code, domain, supportedformats, embed_token)
 168
 169
 170 class OoyalaExternalIE(OoyalaBaseIE):
 171     _VALID_URL = r'''(?x)
 172                     (?:
 173                         ooyalaexternal:|
 174                         https?://.+?\.ooyala\.com/.*?\bexternalId=
 175                     )
 176                     (?P<partner_id>[^:]+)
 177                     :
 178                     (?P<id>.+)
 179                     (?:
 180                         :|
 181                         .*?&pcode=
 182                     )
 183                     (?P<pcode>.+?)
 184                     (?:&|$)
 185                     '''
 186
 187     _TEST = {
 188         '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',
 189         'info_dict': {
 190             'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
 191             'ext': 'mp4',
 192             'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
 193             'duration': 1302.0,
 194         },
 195         'params': {
 196             # m3u8 download
 197             'skip_download': True,
 198         },
 199     }
 200
 201     def _real_extract(self, url):
 202         partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
 203         content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
 204         return self._extract(content_tree_url, video_id)