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

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     compat_urlparse,
   6     compat_urllib_parse,
   7     xpath_with_ns,
   8     determine_ext,
   9 )
  10
  11
  12 class InternetVideoArchiveIE(InfoExtractor):
  13     _VALID_URL = r'https?://video\.internetvideoarchive\.net/flash/players/.*?\?.*?publishedid.*?'
  14
  15     _TEST = {
  16         u'url': u'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?customerid=69249&publishedid=452693&playerid=247',
  17         u'file': u'452693.mp4',
  18         u'info_dict': {
  19             u'title': u'SKYFALL',
  20             u'description': u'In SKYFALL, Bond\'s loyalty to M is tested as her past comes back to haunt her. As MI6 comes under attack, 007 must track down and destroy the threat, no matter how personal the cost.',
  21             u'duration': 153,
  22         },
  23     }
  24
  25     @staticmethod
  26     def _build_url(query):
  27         return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query
  28
  29     @staticmethod
  30     def _clean_query(query):
  31         NEEDED_ARGS = ['publishedid', 'customerid']
  32         query_dic = compat_urlparse.parse_qs(query)
  33         cleaned_dic = dict((k,v[0]) for (k,v) in query_dic.items() if k in NEEDED_ARGS)
  34         # Other player ids return m3u8 urls
  35         cleaned_dic['playerid'] = '247'
  36         cleaned_dic['videokbrate'] = '100000'
  37         return compat_urllib_parse.urlencode(cleaned_dic)
  38
  39     def _real_extract(self, url):
  40         query = compat_urlparse.urlparse(url).query
  41         query_dic = compat_urlparse.parse_qs(query)
  42         video_id = query_dic['publishedid'][0]
  43         url = self._build_url(query)
  44
  45         flashconfiguration = self._download_xml(url, video_id,
  46             u'Downloading flash configuration')
  47         file_url = flashconfiguration.find('file').text
  48         file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx')
  49         # Replace some of the parameters in the query to get the best quality
  50         # and http links (no m3u8 manifests)
  51         file_url = re.sub(r'(?<=\?)(.+)$',
  52             lambda m: self._clean_query(m.group()),
  53             file_url)
  54         info = self._download_xml(file_url, video_id,
  55             u'Downloading video info')
  56         item = info.find('channel/item')
  57
  58         def _bp(p):
  59             return xpath_with_ns(p,
  60                 {'media': 'http://search.yahoo.com/mrss/',
  61                 'jwplayer': 'http://developer.longtailvideo.com/trac/wiki/FlashFormats'})
  62         formats = []
  63         for content in item.findall(_bp('media:group/media:content')):
  64             attr = content.attrib
  65             f_url = attr['url']
  66             formats.append({
  67                 'url': f_url,
  68                 'ext': determine_ext(f_url),
  69                 'width': int(attr['width']),
  70                 'bitrate': int(attr['bitrate']),
  71             })
  72         formats = sorted(formats, key=lambda f: f['bitrate'])
  73
  74         return {
  75             'id': video_id,
  76             'title': item.find('title').text,
  77             'formats': formats,
  78             'thumbnail': item.find(_bp('media:thumbnail')).attrib['url'],
  79             'description': item.find('description').text,
  80             'duration': int(attr['duration']),
  81         }