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

   1 from __future__ import unicode_literals
   2 import re
   3 import json
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     unescapeHTML,
   8     ExtractorError,
   9 )
  10
  11
  12 class OoyalaIE(InfoExtractor):
  13     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
  14
  15     _TESTS = [
  16         {
  17             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
  18             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  19             'md5': '3f5cceb3a7bf461d6c29dc466cf8033c',
  20             'info_dict': {
  21                 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  22                 'ext': 'mp4',
  23                 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
  24                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
  25             },
  26         }, {
  27             # Only available for ipad
  28             'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  29             'md5': '4b9754921fddb68106e48c142e2a01e6',
  30             'info_dict': {
  31                 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  32                 'ext': 'mp4',
  33                 'title': 'Simulation Overview - Levels of Simulation',
  34                 'description': '',
  35             },
  36         },
  37     ]
  38
  39     @staticmethod
  40     def _url_for_embed_code(embed_code):
  41         return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
  42
  43     @classmethod
  44     def _build_url_result(cls, embed_code):
  45         return cls.url_result(cls._url_for_embed_code(embed_code),
  46                               ie=cls.ie_key())
  47
  48     def _extract_result(self, info, more_info):
  49         return {
  50             'id': info['embedCode'],
  51             'ext': 'mp4',
  52             'title': unescapeHTML(info['title']),
  53             'url': info.get('ipad_url') or info['url'],
  54             'description': unescapeHTML(more_info['description']),
  55             'thumbnail': more_info['promo'],
  56         }
  57
  58     def _real_extract(self, url):
  59         mobj = re.match(self._VALID_URL, url)
  60         embedCode = mobj.group('id')
  61         player_url = 'http://player.ooyala.com/player.js?embedCode=%s' % embedCode
  62         player = self._download_webpage(player_url, embedCode)
  63         mobile_url = self._search_regex(r'mobile_player_url="(.+?)&device="',
  64                                         player, 'mobile player url')
  65         # Looks like some videos are only available for particular devices
  66         # (e.g. http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0
  67         # is only available for ipad)
  68         # Working around with fetching URLs for all the devices found starting with 'unknown'
  69         # until we succeed or eventually fail for each device.
  70         devices = re.findall(r'device\s*=\s*"([^"]+)";', player)
  71         devices.remove('unknown')
  72         devices.insert(0, 'unknown')
  73         for device in devices:
  74             mobile_player = self._download_webpage(
  75                 '%s&device=%s' % (mobile_url, device), embedCode,
  76                 'Downloading mobile player JS for %s device' % device)
  77             videos_info = self._search_regex(
  78                 r'var streams=window.oo_testEnv\?\[\]:eval\("\((\[{.*?}\])\)"\);',
  79                 mobile_player, 'info', fatal=False, default=None)
  80             if videos_info:
  81                 break
  82         if not videos_info:
  83             raise ExtractorError('Unable to extract info')
  84         videos_info = videos_info.replace('\\"', '"')
  85         videos_more_info = self._search_regex(
  86             r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, 'more info').replace('\\"', '"')
  87         videos_info = json.loads(videos_info)
  88         videos_more_info = json.loads(videos_more_info)
  89
  90         if videos_more_info.get('lineup'):
  91             videos = [self._extract_result(info, more_info) for (info, more_info) in zip(videos_info, videos_more_info['lineup'])]
  92             return {
  93                 '_type': 'playlist',
  94                 'id': embedCode,
  95                 'title': unescapeHTML(videos_more_info['title']),
  96                 'entries': videos,
  97             }
  98         else:
  99             return self._extract_result(videos_info[0], videos_more_info)