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