Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ooyala.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_b64decode,
   8     compat_str,
   9     compat_urllib_parse_urlencode,
  10 )
  11 from ..utils import (
  12     determine_ext,
  13     ExtractorError,
  14     float_or_none,
  15     int_or_none,
  16     try_get,
  17     unsmuggle_url,
  18 )
  19
  20
  21 class OoyalaBaseIE(InfoExtractor):
  22     _PLAYER_BASE = 'http://player.ooyala.com/'
  23     _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
  24     _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s?'
  25
  26     def _extract(self, content_tree_url, video_id, domain='example.org', supportedformats=None, embed_token=None):
  27         content_tree = self._download_json(content_tree_url, video_id)['content_tree']
  28         metadata = content_tree[list(content_tree)[0]]
  29         embed_code = metadata['embed_code']
  30         pcode = metadata.get('asset_pcode') or embed_code
  31         title = metadata['title']
  32
  33         auth_data = self._download_json(
  34             self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) +
  35             compat_urllib_parse_urlencode({
  36                 'domain': domain,
  37                 'supportedFormats': supportedformats or 'mp4,rtmp,m3u8,hds,dash,smooth',
  38                 'embedToken': embed_token,
  39             }), video_id)
  40
  41         cur_auth_data = auth_data['authorization_data'][embed_code]
  42
  43         urls = []
  44         formats = []
  45         if cur_auth_data['authorized']:
  46             for stream in cur_auth_data['streams']:
  47                 url_data = try_get(stream, lambda x: x['url']['data'], compat_str)
  48                 if not url_data:
  49                     continue
  50                 s_url = compat_b64decode(url_data).decode('utf-8')
  51                 if not s_url or s_url in urls:
  52                     continue
  53                 urls.append(s_url)
  54                 ext = determine_ext(s_url, None)
  55                 delivery_type = stream.get('delivery_type')
  56                 if delivery_type == 'hls' or ext == 'm3u8':
  57                     formats.extend(self._extract_m3u8_formats(
  58                         re.sub(r'/ip(?:ad|hone)/', '/all/', s_url), embed_code, 'mp4', 'm3u8_native',
  59                         m3u8_id='hls', fatal=False))
  60                 elif delivery_type == 'hds' or ext == 'f4m':
  61                     formats.extend(self._extract_f4m_formats(
  62                         s_url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
  63                 elif delivery_type == 'dash' or ext == 'mpd':
  64                     formats.extend(self._extract_mpd_formats(
  65                         s_url, embed_code, mpd_id='dash', fatal=False))
  66                 elif delivery_type == 'smooth':
  67                     self._extract_ism_formats(
  68                         s_url, embed_code, ism_id='mss', fatal=False)
  69                 elif ext == 'smil':
  70                     formats.extend(self._extract_smil_formats(
  71                         s_url, embed_code, fatal=False))
  72                 else:
  73                     formats.append({
  74                         'url': s_url,
  75                         'ext': ext or delivery_type,
  76                         'vcodec': stream.get('video_codec'),
  77                         'format_id': delivery_type,
  78                         'width': int_or_none(stream.get('width')),
  79                         'height': int_or_none(stream.get('height')),
  80                         'abr': int_or_none(stream.get('audio_bitrate')),
  81                         'vbr': int_or_none(stream.get('video_bitrate')),
  82                         'fps': float_or_none(stream.get('framerate')),
  83                     })
  84         else:
  85             raise ExtractorError('%s said: %s' % (
  86                 self.IE_NAME, cur_auth_data['message']), expected=True)
  87         self._sort_formats(formats)
  88
  89         subtitles = {}
  90         for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items():
  91             sub_url = sub.get('url')
  92             if not sub_url:
  93                 continue
  94             subtitles[lang] = [{
  95                 'url': sub_url,
  96             }]
  97
  98         return {
  99             'id': embed_code,
 100             'title': title,
 101             'description': metadata.get('description'),
 102             'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
 103             'duration': float_or_none(metadata.get('duration'), 1000),
 104             'subtitles': subtitles,
 105             'formats': formats,
 106         }
 107
 108
 109 class OoyalaIE(OoyalaBaseIE):
 110     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
 111
 112     _TESTS = [
 113         {
 114             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
 115             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
 116             'info_dict': {
 117                 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
 118                 'ext': 'mp4',
 119                 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
 120                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
 121                 'duration': 853.386,
 122             },
 123             # The video in the original webpage now uses PlayWire
 124             'skip': 'Ooyala said: movie expired',
 125         }, {
 126             # Only available for ipad
 127             'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 128             'info_dict': {
 129                 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 130                 'ext': 'mp4',
 131                 'title': 'Simulation Overview - Levels of Simulation',
 132                 'duration': 194.948,
 133             },
 134         },
 135         {
 136             # Information available only through SAS api
 137             # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
 138             'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 139             'md5': 'a84001441b35ea492bc03736e59e7935',
 140             'info_dict': {
 141                 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 142                 'ext': 'mp4',
 143                 'title': 'Divide Tool Path.mp4',
 144                 'duration': 204.405,
 145             }
 146         },
 147         {
 148             # empty stream['url']['data']
 149             'url': 'http://player.ooyala.com/player.js?embedCode=w2bnZtYjE6axZ_dw1Cd0hQtXd_ige2Is',
 150             'only_matching': True,
 151         }
 152     ]
 153
 154     @staticmethod
 155     def _url_for_embed_code(embed_code):
 156         return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
 157
 158     @classmethod
 159     def _build_url_result(cls, embed_code):
 160         return cls.url_result(cls._url_for_embed_code(embed_code),
 161                               ie=cls.ie_key())
 162
 163     def _real_extract(self, url):
 164         url, smuggled_data = unsmuggle_url(url, {})
 165         embed_code = self._match_id(url)
 166         domain = smuggled_data.get('domain')
 167         supportedformats = smuggled_data.get('supportedformats')
 168         embed_token = smuggled_data.get('embed_token')
 169         content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
 170         return self._extract(content_tree_url, embed_code, domain, supportedformats, embed_token)
 171
 172
 173 class OoyalaExternalIE(OoyalaBaseIE):
 174     _VALID_URL = r'''(?x)
 175                     (?:
 176                         ooyalaexternal:|
 177                         https?://.+?\.ooyala\.com/.*?\bexternalId=
 178                     )
 179                     (?P<partner_id>[^:]+)
 180                     :
 181                     (?P<id>.+)
 182                     (?:
 183                         :|
 184                         .*?&pcode=
 185                     )
 186                     (?P<pcode>.+?)
 187                     (?:&|$)
 188                     '''
 189
 190     _TEST = {
 191         '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',
 192         'info_dict': {
 193             'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
 194             'ext': 'mp4',
 195             'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
 196             'duration': 1302.0,
 197         },
 198         'params': {
 199             # m3u8 download
 200             'skip_download': True,
 201         },
 202     }
 203
 204     def _real_extract(self, url):
 205         partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
 206         content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
 207         return self._extract(content_tree_url, video_id)