_ Git - youtube-dl/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     determine_ext,
  10 )
  11
  12
  13 class OoyalaIE(InfoExtractor):
  14     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
  15
  16     _TESTS = [
  17         {
  18             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
  19             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  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             'info_dict': {
  30                 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  31                 'ext': 'mp4',
  32                 'title': 'Simulation Overview - Levels of Simulation',
  33                 'description': '',
  34             },
  35         },
  36     ]
  37
  38     @staticmethod
  39     def _url_for_embed_code(embed_code):
  40         return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
  41
  42     @classmethod
  43     def _build_url_result(cls, embed_code):
  44         return cls.url_result(cls._url_for_embed_code(embed_code),
  45                               ie=cls.ie_key())
  46
  47     def _extract_result(self, info, more_info):
  48         embedCode = info['embedCode']
  49         video_url = info.get('ipad_url') or info['url']
  50
  51         if determine_ext(video_url) == 'm3u8':
  52             formats = self._extract_m3u8_formats(video_url, embedCode, ext='mp4')
  53         else:
  54             formats = [{
  55                 'url': video_url,
  56                 'ext': 'mp4',
  57             }]
  58
  59         return {
  60             'id': embedCode,
  61             'title': unescapeHTML(info['title']),
  62             'formats': formats,
  63             'description': unescapeHTML(more_info['description']),
  64             'thumbnail': more_info['promo'],
  65         }
  66
  67     def _real_extract(self, url):
  68         mobj = re.match(self._VALID_URL, url)
  69         embedCode = mobj.group('id')
  70         player_url = 'http://player.ooyala.com/player.js?embedCode=%s' % embedCode
  71         player = self._download_webpage(player_url, embedCode)
  72         mobile_url = self._search_regex(r'mobile_player_url="(.+?)&device="',
  73                                         player, 'mobile player url')
  74         # Looks like some videos are only available for particular devices
  75         # (e.g. http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0
  76         # is only available for ipad)
  77         # Working around with fetching URLs for all the devices found starting with 'unknown'
  78         # until we succeed or eventually fail for each device.
  79         devices = re.findall(r'device\s*=\s*"([^"]+)";', player)
  80         devices.remove('unknown')
  81         devices.insert(0, 'unknown')
  82         for device in devices:
  83             mobile_player = self._download_webpage(
  84                 '%s&device=%s' % (mobile_url, device), embedCode,
  85                 'Downloading mobile player JS for %s device' % device)
  86             videos_info = self._search_regex(
  87                 r'var streams=window.oo_testEnv\?\[\]:eval\("\((\[{.*?}\])\)"\);',
  88                 mobile_player, 'info', fatal=False, default=None)
  89             if videos_info:
  90                 break
  91         if not videos_info:
  92             raise ExtractorError('Unable to extract info')
  93         videos_info = videos_info.replace('\\"', '"')
  94         videos_more_info = self._search_regex(
  95             r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, 'more info').replace('\\"', '"')
  96         videos_info = json.loads(videos_info)
  97         videos_more_info = json.loads(videos_more_info)
  98
  99         if videos_more_info.get('lineup'):
 100             videos = [self._extract_result(info, more_info) for (info, more_info) in zip(videos_info, videos_more_info['lineup'])]
 101             return {
 102                 '_type': 'playlist',
 103                 'id': embedCode,
 104                 'title': unescapeHTML(videos_more_info['title']),
 105                 'entries': videos,
 106             }
 107         else:
 108             return self._extract_result(videos_info[0], videos_more_info)