_ Git - youtube-dl/blob - youtube_dl/extractor/trilulilu.py

   1 import json
   2 import re
   3 import xml.etree.ElementTree
   4
   5 from .common import InfoExtractor
   6
   7
   8 class TriluliluIE(InfoExtractor):
   9     _VALID_URL = r'(?x)(?:https?://)?(?:www\.)?trilulilu\.ro/video-(?P<category>[^/]+)/(?P<video_id>[^/]+)'
  10     _TEST = {
  11         u"url": u"http://www.trilulilu.ro/video-animatie/big-buck-bunny-1",
  12         u'file': u"big-buck-bunny-1.mp4",
  13         u'info_dict': {
  14             u"title": u"Big Buck Bunny",
  15             u"description": u":) pentru copilul din noi",
  16         },
  17         # Server ignores Range headers (--test)
  18         u"params": {
  19             u"skip_download": True
  20         }
  21     }
  22
  23     def _real_extract(self, url):
  24         mobj = re.match(self._VALID_URL, url)
  25         video_id = mobj.group('video_id')
  26
  27         webpage = self._download_webpage(url, video_id)
  28
  29         title = self._og_search_title(webpage)
  30         thumbnail = self._og_search_thumbnail(webpage)
  31         description = self._og_search_description(webpage)
  32
  33         log_str = self._search_regex(
  34             r'block_flash_vars[ ]=[ ]({[^}]+})', webpage, u'log info')
  35         log = json.loads(log_str)
  36
  37         format_url = (u'http://fs%(server)s.trilulilu.ro/%(hash)s/'
  38                       u'video-formats2' % log)
  39         format_str = self._download_webpage(
  40             format_url, video_id,
  41             note=u'Downloading formats',
  42             errnote=u'Error while downloading formats')
  43
  44         format_doc = xml.etree.ElementTree.fromstring(format_str)
  45
  46         video_url_template = (
  47             u'http://fs%(server)s.trilulilu.ro/stream.php?type=video'
  48             u'&source=site&hash=%(hash)s&username=%(userid)s&'
  49             u'key=ministhebest&format=%%s&sig=&exp=' %
  50             log)
  51         formats = [
  52             {
  53                 'format': fnode.text,
  54                 'url': video_url_template % fnode.text,
  55             }
  56
  57             for fnode in format_doc.findall('./formats/format')
  58         ]
  59
  60         info = {
  61             '_type': 'video',
  62             'id': video_id,
  63             'formats': formats,
  64             'title': title,
  65             'description': description,
  66             'thumbnail': thumbnail,
  67         }
  68
  69         # TODO: Remove when #980 has been merged
  70         info['url'] = formats[-1]['url']
  71         info['ext'] = formats[-1]['format'].partition('-')[0]
  72
  73         return info