Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/mtv.py

   1 import re
   2 import socket
   3 import xml.etree.ElementTree
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_http_client,
   8     compat_str,
   9     compat_urllib_error,
  10     compat_urllib_request,
  11
  12     ExtractorError,
  13 )
  14
  15
  16 class MTVIE(InfoExtractor):
  17     _VALID_URL = r'^(?P<proto>https?://)?(?:www\.)?mtv\.com/videos/[^/]+/(?P<videoid>[0-9]+)/[^/]+$'
  18     _WORKING = False
  19
  20     def _real_extract(self, url):
  21         mobj = re.match(self._VALID_URL, url)
  22         if mobj is None:
  23             raise ExtractorError(u'Invalid URL: %s' % url)
  24         if not mobj.group('proto'):
  25             url = 'http://' + url
  26         video_id = mobj.group('videoid')
  27
  28         webpage = self._download_webpage(url, video_id)
  29
  30         # Some videos come from Vevo.com
  31         m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";',
  32                            webpage, re.DOTALL)
  33         if m_vevo:
  34             vevo_id = m_vevo.group(1);
  35             self.to_screen(u'Vevo video detected: %s' % vevo_id)
  36             return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
  37
  38         #song_name = self._html_search_regex(r'<meta name="mtv_vt" content="([^"]+)"/>',
  39         #    webpage, u'song name', fatal=False)
  40
  41         video_title = self._html_search_regex(r'<meta name="mtv_an" content="([^"]+)"/>',
  42             webpage, u'title')
  43
  44         mtvn_uri = self._html_search_regex(r'<meta name="mtvn_uri" content="([^"]+)"/>',
  45             webpage, u'mtvn_uri', fatal=False)
  46
  47         content_id = self._search_regex(r'MTVN.Player.defaultPlaylistId = ([0-9]+);',
  48             webpage, u'content id', fatal=False)
  49
  50         videogen_url = 'http://www.mtv.com/player/includes/mediaGen.jhtml?uri=' + mtvn_uri + '&id=' + content_id + '&vid=' + video_id + '&ref=www.mtvn.com&viewUri=' + mtvn_uri
  51         self.report_extraction(video_id)
  52         request = compat_urllib_request.Request(videogen_url)
  53         try:
  54             metadataXml = compat_urllib_request.urlopen(request).read()
  55         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  56             raise ExtractorError(u'Unable to download video metadata: %s' % compat_str(err))
  57
  58         mdoc = xml.etree.ElementTree.fromstring(metadataXml)
  59         renditions = mdoc.findall('.//rendition')
  60
  61         # For now, always pick the highest quality.
  62         rendition = renditions[-1]
  63
  64         try:
  65             _,_,ext = rendition.attrib['type'].partition('/')
  66             format = ext + '-' + rendition.attrib['width'] + 'x' + rendition.attrib['height'] + '_' + rendition.attrib['bitrate']
  67             video_url = rendition.find('./src').text
  68         except KeyError:
  69             raise ExtractorError('Invalid rendition field.')
  70
  71         info = {
  72             'id': video_id,
  73             'url': video_url,
  74             'upload_date': None,
  75             'title': video_title,
  76             'ext': ext,
  77             'format': format,
  78         }
  79
  80         return [info]