yt_dlp/extractor/ooyala.py

   1 from __future__ import unicode_literals
   2
   3 import base64
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_b64decode,
   9     compat_str,
  10 )
  11 from ..utils import (
  12     determine_ext,
  13     float_or_none,
  14     int_or_none,
  15     try_get,
  16     unsmuggle_url,
  17 )
  18
  19
  20 class OoyalaBaseIE(InfoExtractor):
  21     _PLAYER_BASE = 'http://player.ooyala.com/'
  22     _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
  23     _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s'
  24
  25     def _extract(self, content_tree_url, video_id, domain=None, supportedformats=None, embed_token=None):
  26         content_tree = self._download_json(content_tree_url, video_id)['content_tree']
  27         metadata = content_tree[list(content_tree)[0]]
  28         embed_code = metadata['embed_code']
  29         pcode = metadata.get('asset_pcode') or embed_code
  30         title = metadata['title']
  31
  32         auth_data = self._download_json(
  33             self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code),
  34             video_id, headers=self.geo_verification_headers(), query={
  35                 'domain': domain or 'player.ooyala.com',
  36                 'supportedFormats': supportedformats or 'mp4,rtmp,m3u8,hds,dash,smooth',
  37                 'embedToken': embed_token,
  38             })['authorization_data'][embed_code]
  39
  40         urls = []
  41         formats = []
  42         streams = auth_data.get('streams') or [{
  43             'delivery_type': 'hls',
  44             'url': {
  45                 'data': base64.b64encode(('http://player.ooyala.com/hls/player/all/%s.m3u8' % embed_code).encode()).decode(),
  46             }
  47         }]
  48         for stream in streams:
  49             url_data = try_get(stream, lambda x: x['url']['data'], compat_str)
  50             if not url_data:
  51                 continue
  52             s_url = compat_b64decode(url_data).decode('utf-8')
  53             if not s_url or s_url in urls:
  54                 continue
  55             urls.append(s_url)
  56             ext = determine_ext(s_url, None)
  57             delivery_type = stream.get('delivery_type')
  58             if delivery_type == 'hls' or ext == 'm3u8':
  59                 formats.extend(self._extract_m3u8_formats(
  60                     re.sub(r'/ip(?:ad|hone)/', '/all/', s_url), embed_code, 'mp4', 'm3u8_native',
  61                     m3u8_id='hls', fatal=False))
  62             elif delivery_type == 'hds' or ext == 'f4m':
  63                 formats.extend(self._extract_f4m_formats(
  64                     s_url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
  65             elif delivery_type == 'dash' or ext == 'mpd':
  66                 formats.extend(self._extract_mpd_formats(
  67                     s_url, embed_code, mpd_id='dash', fatal=False))
  68             elif delivery_type == 'smooth':
  69                 self._extract_ism_formats(
  70                     s_url, embed_code, ism_id='mss', fatal=False)
  71             elif ext == 'smil':
  72                 formats.extend(self._extract_smil_formats(
  73                     s_url, embed_code, fatal=False))
  74             else:
  75                 formats.append({
  76                     'url': s_url,
  77                     'ext': ext or delivery_type,
  78                     'vcodec': stream.get('video_codec'),
  79                     'format_id': delivery_type,
  80                     'width': int_or_none(stream.get('width')),
  81                     'height': int_or_none(stream.get('height')),
  82                     'abr': int_or_none(stream.get('audio_bitrate')),
  83                     'vbr': int_or_none(stream.get('video_bitrate')),
  84                     'fps': float_or_none(stream.get('framerate')),
  85                 })
  86         if not formats and not auth_data.get('authorized'):
  87             self.raise_no_formats('%s said: %s' % (
  88                 self.IE_NAME, auth_data['message']), expected=True)
  89         self._sort_formats(formats)
  90
  91         subtitles = {}
  92         for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items():
  93             sub_url = sub.get('url')
  94             if not sub_url:
  95                 continue
  96             subtitles[lang] = [{
  97                 'url': sub_url,
  98             }]
  99
 100         return {
 101             'id': embed_code,
 102             'title': title,
 103             'description': metadata.get('description'),
 104             'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
 105             'duration': float_or_none(metadata.get('duration'), 1000),
 106             'subtitles': subtitles,
 107             'formats': formats,
 108         }
 109
 110
 111 class OoyalaIE(OoyalaBaseIE):
 112     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
 113
 114     _TESTS = [
 115         {
 116             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
 117             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
 118             'info_dict': {
 119                 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
 120                 'ext': 'mp4',
 121                 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
 122                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
 123                 'duration': 853.386,
 124             },
 125             # The video in the original webpage now uses PlayWire
 126             'skip': 'Ooyala said: movie expired',
 127         }, {
 128             # Only available for ipad
 129             'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 130             'info_dict': {
 131                 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
 132                 'ext': 'mp4',
 133                 'title': 'Simulation Overview - Levels of Simulation',
 134                 'duration': 194.948,
 135             },
 136         },
 137         {
 138             # Information available only through SAS api
 139             # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
 140             'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 141             'md5': 'a84001441b35ea492bc03736e59e7935',
 142             'info_dict': {
 143                 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
 144                 'ext': 'mp4',
 145                 'title': 'Divide Tool Path.mp4',
 146                 'duration': 204.405,
 147             }
 148         },
 149         {
 150             # empty stream['url']['data']
 151             'url': 'http://player.ooyala.com/player.js?embedCode=w2bnZtYjE6axZ_dw1Cd0hQtXd_ige2Is',
 152             'only_matching': True,
 153         }
 154     ]
 155
 156     @staticmethod
 157     def _url_for_embed_code(embed_code):
 158         return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
 159
 160     @classmethod
 161     def _build_url_result(cls, embed_code):
 162         return cls.url_result(cls._url_for_embed_code(embed_code),
 163                               ie=cls.ie_key())
 164
 165     def _real_extract(self, url):
 166         url, smuggled_data = unsmuggle_url(url, {})
 167         embed_code = self._match_id(url)
 168         domain = smuggled_data.get('domain')
 169         supportedformats = smuggled_data.get('supportedformats')
 170         embed_token = smuggled_data.get('embed_token')
 171         content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
 172         return self._extract(content_tree_url, embed_code, domain, supportedformats, embed_token)
 173
 174
 175 class OoyalaExternalIE(OoyalaBaseIE):
 176     _VALID_URL = r'''(?x)
 177                     (?:
 178                         ooyalaexternal:|
 179                         https?://.+?\.ooyala\.com/.*?\bexternalId=
 180                     )
 181                     (?P<partner_id>[^:]+)
 182                     :
 183                     (?P<id>.+)
 184                     (?:
 185                         :|
 186                         .*?&pcode=
 187                     )
 188                     (?P<pcode>.+?)
 189                     (?:&|$)
 190                     '''
 191
 192     _TEST = {
 193         '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',
 194         'info_dict': {
 195             'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
 196             'ext': 'mp4',
 197             'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
 198             'duration': 1302.0,
 199         },
 200         'params': {
 201             # m3u8 download
 202             'skip_download': True,
 203         },
 204     }
 205
 206     def _real_extract(self, url):
 207         partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
 208         content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
 209         return self._extract(content_tree_url, video_id)