youtube_dl/extractor/internetvideoarchive.py

   1 import re
   2 import xml.etree.ElementTree
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_urlparse,
   7     compat_urllib_parse,
   8     xpath_with_ns,
   9     determine_ext,
  10 )
  11
  12
  13 class InternetVideoArchiveIE(InfoExtractor):
  14     _VALID_URL = r'https?://video\.internetvideoarchive\.net/flash/players/.*?\?.*?publishedid.*?'
  15
  16     _TEST = {
  17         u'url': u'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?customerid=69249&publishedid=452693&playerid=247',
  18         u'file': u'452693.mp4',
  19         u'info_dict': {
  20             u'title': u'SKYFALL',
  21             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.',
  22             u'duration': 153,
  23         },
  24     }
  25
  26     @staticmethod
  27     def _build_url(query):
  28         return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query
  29
  30     @staticmethod
  31     def _clean_query(query):
  32         NEEDED_ARGS = ['publishedid', 'customerid']
  33         query_dic = compat_urlparse.parse_qs(query)
  34         cleaned_dic = dict((k,v[0]) for (k,v) in query_dic.items() if k in NEEDED_ARGS)
  35         # Other player ids return m3u8 urls
  36         cleaned_dic['playerid'] = '247'
  37         cleaned_dic['videokbrate'] = '100000'
  38         return compat_urllib_parse.urlencode(cleaned_dic)
  39
  40     def _real_extract(self, url):
  41         query = compat_urlparse.urlparse(url).query
  42         query_dic = compat_urlparse.parse_qs(query)
  43         video_id = query_dic['publishedid'][0]
  44         url = self._build_url(query)
  45
  46         flashconfiguration_xml = self._download_webpage(url, video_id,
  47             u'Downloading flash configuration')
  48         flashconfiguration = xml.etree.ElementTree.fromstring(flashconfiguration_xml.encode('utf-8'))
  49         file_url = flashconfiguration.find('file').text
  50         file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx')
  51         # Replace some of the parameters in the query to get the best quality
  52         # and http links (no m3u8 manifests)
  53         file_url = re.sub(r'(?<=\?)(.+)$',
  54             lambda m: self._clean_query(m.group()),
  55             file_url)
  56         info_xml = self._download_webpage(file_url, video_id,
  57             u'Downloading video info')
  58         info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
  59         item = info.find('channel/item')
  60
  61         def _bp(p):
  62             return xpath_with_ns(p,
  63                 {'media': 'http://search.yahoo.com/mrss/',
  64                 'jwplayer': 'http://developer.longtailvideo.com/trac/wiki/FlashFormats'})
  65         formats = []
  66         for content in item.findall(_bp('media:group/media:content')):
  67             attr = content.attrib
  68             f_url = attr['url']
  69             formats.append({
  70                 'url': f_url,
  71                 'ext': determine_ext(f_url),
  72                 'width': int(attr['width']),
  73                 'bitrate': int(attr['bitrate']),
  74             })
  75         formats = sorted(formats, key=lambda f: f['bitrate'])
  76
  77         return {
  78             'id': video_id,
  79             'title': item.find('title').text,
  80             'formats': formats,
  81             'thumbnail': item.find(_bp('media:thumbnail')).attrib['url'],
  82             'description': item.find('description').text,
  83             'duration': int(attr['duration']),
  84         }