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