_ Git - youtube-dl/blob - 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     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': 156,
  22         },
  23     }
  24
  25     @staticmethod
  26     def _build_url(query):
  27         return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query
  28
  29     def _real_extract(self, url):
  30         query = compat_urlparse.urlparse(url).query
  31         query_dic = compat_urlparse.parse_qs(query)
  32         video_id = query_dic['publishedid'][0]
  33         url = self._build_url(query)
  34
  35         flashconfiguration_xml = self._download_webpage(url, video_id,
  36             u'Downloading flash configuration')
  37         flashconfiguration = xml.etree.ElementTree.fromstring(flashconfiguration_xml.encode('utf-8'))
  38         file_url = flashconfiguration.find('file').text
  39         file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx')
  40         info_xml = self._download_webpage(file_url, video_id,
  41             u'Downloading video info')
  42         info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
  43         item = info.find('channel/item')
  44
  45         def _bp(p):
  46             return xpath_with_ns(p,
  47                 {'media': 'http://search.yahoo.com/mrss/',
  48                 'jwplayer': 'http://developer.longtailvideo.com/trac/wiki/FlashFormats'})
  49         formats = []
  50         for content in item.findall(_bp('media:group/media:content')):
  51             attr = content.attrib
  52             f_url = attr['url']
  53             formats.append({
  54                 'url': f_url,
  55                 'ext': determine_ext(f_url),
  56                 'width': int(attr['width']),
  57                 'bitrate': int(attr['bitrate']),
  58             })
  59         formats = sorted(formats, key=lambda f: f['bitrate'])
  60
  61         info = {
  62             'id': video_id,
  63             'title': item.find('title').text,
  64             'formats': formats,
  65             'thumbnail': item.find(_bp('media:thumbnail')).attrib['url'],
  66             'description': item.find('description').text,
  67             'duration': int(attr['duration']),
  68         }
  69         # TODO: Remove when #980 has been merged
  70         info.update(formats[-1])
  71         return info