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

   1 import re
   2 import socket
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_http_client,
   7     compat_parse_qs,
   8     compat_urllib_error,
   9     compat_urllib_parse,
  10     compat_urllib_request,
  11     compat_str,
  12
  13     ExtractorError,
  14 )
  15
  16 class MetacafeIE(InfoExtractor):
  17     """Information Extractor for metacafe.com."""
  18
  19     _VALID_URL = r'(?:http://)?(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*'
  20     _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
  21     _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
  22     IE_NAME = u'metacafe'
  23
  24     def report_disclaimer(self):
  25         """Report disclaimer retrieval."""
  26         self.to_screen(u'Retrieving disclaimer')
  27
  28     def _real_initialize(self):
  29         # Retrieve disclaimer
  30         request = compat_urllib_request.Request(self._DISCLAIMER)
  31         try:
  32             self.report_disclaimer()
  33             compat_urllib_request.urlopen(request).read()
  34         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  35             raise ExtractorError(u'Unable to retrieve disclaimer: %s' % compat_str(err))
  36
  37         # Confirm age
  38         disclaimer_form = {
  39             'filters': '0',
  40             'submit': "Continue - I'm over 18",
  41             }
  42         request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form))
  43         try:
  44             self.report_age_confirmation()
  45             compat_urllib_request.urlopen(request).read()
  46         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  47             raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
  48
  49     def _real_extract(self, url):
  50         # Extract id and simplified title from URL
  51         mobj = re.match(self._VALID_URL, url)
  52         if mobj is None:
  53             raise ExtractorError(u'Invalid URL: %s' % url)
  54
  55         video_id = mobj.group(1)
  56
  57         # Check if video comes from YouTube
  58         mobj2 = re.match(r'^yt-(.*)$', video_id)
  59         if mobj2 is not None:
  60             return [self.url_result('http://www.youtube.com/watch?v=%s' % mobj2.group(1), 'Youtube')]
  61
  62         # Retrieve video webpage to extract further information
  63         webpage = self._download_webpage('http://www.metacafe.com/watch/%s/' % video_id, video_id)
  64
  65         # Extract URL, uploader and title from webpage
  66         self.report_extraction(video_id)
  67         mobj = re.search(r'(?m)&mediaURL=([^&]+)', webpage)
  68         if mobj is not None:
  69             mediaURL = compat_urllib_parse.unquote(mobj.group(1))
  70             video_extension = mediaURL[-3:]
  71
  72             # Extract gdaKey if available
  73             mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage)
  74             if mobj is None:
  75                 video_url = mediaURL
  76             else:
  77                 gdaKey = mobj.group(1)
  78                 video_url = '%s?__gda__=%s' % (mediaURL, gdaKey)
  79         else:
  80             mobj = re.search(r' name="flashvars" value="(.*?)"', webpage)
  81             if mobj is None:
  82                 raise ExtractorError(u'Unable to extract media URL')
  83             vardict = compat_parse_qs(mobj.group(1))
  84             if 'mediaData' not in vardict:
  85                 raise ExtractorError(u'Unable to extract media URL')
  86             mobj = re.search(r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
  87             if mobj is None:
  88                 raise ExtractorError(u'Unable to extract media URL')
  89             mediaURL = mobj.group('mediaURL').replace('\\/', '/')
  90             video_extension = mediaURL[-3:]
  91             video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
  92
  93         mobj = re.search(r'(?im)<title>(.*) - Video</title>', webpage)
  94         if mobj is None:
  95             raise ExtractorError(u'Unable to extract title')
  96         video_title = mobj.group(1).decode('utf-8')
  97
  98         mobj = re.search(r'submitter=(.*?);', webpage)
  99         if mobj is None:
 100             raise ExtractorError(u'Unable to extract uploader nickname')
 101         video_uploader = mobj.group(1)
 102
 103         return [{
 104             'id':       video_id.decode('utf-8'),
 105             'url':      video_url.decode('utf-8'),
 106             'uploader': video_uploader.decode('utf-8'),
 107             'upload_date':  None,
 108             'title':    video_title,
 109             'ext':      video_extension.decode('utf-8'),
 110         }]