+ if format_param == '0':
+ format_param = self._available_formats[quality_index]
+ best_quality = True
+
+ while True:
+ # Extension
+ video_extension = self._video_extensions.get(format_param, 'flv')
+
+ # Get video info
+ video_info_url = 'http://www.youtube.com/get_video_info?&video_id=%s&el=detailpage&ps=default&eurl=&gl=US&hl=en' % video_id
+ request = urllib2.Request(video_info_url, None, std_headers)
+ try:
+ self.report_video_info_webpage_download(video_id)
+ video_info_webpage = urllib2.urlopen(request).read()
+ except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+ self._downloader.trouble(u'ERROR: unable to download video info webpage: %s' % str(err))
+ return
+ self.report_information_extraction(video_id)
+
+ # "t" param
+ mobj = re.search(r'(?m)&token=([^&]+)(?:&|$)', video_info_webpage)
+ if mobj is None:
+ # Attempt to see if YouTube has issued an error message
+ mobj = re.search(r'(?m)&reason=([^&]+)(?:&|$)', video_info_webpage)
+ if mobj is None:
+ self._downloader.trouble(u'ERROR: unable to extract "t" parameter for unknown reason')
+ stream = open('reportme-ydl-%s.dat' % time.time(), 'wb')
+ stream.write(video_info_webpage)
+ stream.close()
+ else:
+ reason = urllib.unquote_plus(mobj.group(1))
+ self._downloader.trouble(u'ERROR: YouTube said: %s' % reason.decode('utf-8'))
+ return
+ token = urllib.unquote(mobj.group(1))
+ video_real_url = 'http://www.youtube.com/get_video?video_id=%s&t=%s&eurl=&el=detailpage&ps=default&gl=US&hl=en' % (video_id, token)
+ if format_param is not None:
+ video_real_url = '%s&fmt=%s' % (video_real_url, format_param)
+
+ # uploader
+ mobj = re.search(r'(?m)&author=([^&]+)(?:&|$)', video_info_webpage)
+ if mobj is None:
+ self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
+ return
+ video_uploader = urllib.unquote(mobj.group(1))
+
+ # title
+ mobj = re.search(r'(?m)&title=([^&]*)(?:&|$)', video_info_webpage)
+ if mobj is None:
+ self._downloader.trouble(u'ERROR: unable to extract video title')
+ return
+ video_title = urllib.unquote(mobj.group(1))
+ video_title = video_title.decode('utf-8')
+ video_title = re.sub(ur'(?u)&(.+?);', self.htmlentity_transform, video_title)
+ video_title = video_title.replace(os.sep, u'%')
+
+ # simplified title
+ simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
+ simple_title = simple_title.strip(ur'_')
+
+ try:
+ # Process video information
+ self._downloader.process_info({
+ 'id': video_id.decode('utf-8'),
+ 'url': video_real_url.decode('utf-8'),
+ 'uploader': video_uploader.decode('utf-8'),
+ 'title': video_title,
+ 'stitle': simple_title,
+ 'ext': video_extension.decode('utf-8'),
+ })
+
+ return
+
+ except UnavailableFormatError, err:
+ if best_quality:
+ if quality_index == len(self._available_formats) - 1:
+ # I don't ever expect this to happen
+ self._downloader.trouble(u'ERROR: no known formats available for video')
+ return
+ else:
+ self.report_unavailable_format(video_id, format_param)
+ quality_index += 1
+ format_param = self._available_formats[quality_index]
+ continue
+ else:
+ self._downloader.trouble('ERROR: format not available for video')
+ return
+
+
+class MetacafeIE(InfoExtractor):
+ """Information Extractor for metacafe.com."""
+
+ _VALID_URL = r'(?:http://)?(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*'
+ _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
+ _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
+ _youtube_ie = None
+
+ def __init__(self, youtube_ie, downloader=None):
+ InfoExtractor.__init__(self, downloader)
+ self._youtube_ie = youtube_ie
+
+ @staticmethod
+ def suitable(url):
+ return (re.match(MetacafeIE._VALID_URL, url) is not None)
+
+ def report_disclaimer(self):
+ """Report disclaimer retrieval."""
+ self._downloader.to_stdout(u'[metacafe] Retrieving disclaimer')
+
+ def report_age_confirmation(self):
+ """Report attempt to confirm age."""
+ self._downloader.to_stdout(u'[metacafe] Confirming age')
+
+ def report_download_webpage(self, video_id):
+ """Report webpage download."""
+ self._downloader.to_stdout(u'[metacafe] %s: Downloading webpage' % video_id)
+
+ def report_extraction(self, video_id):
+ """Report information extraction."""
+ self._downloader.to_stdout(u'[metacafe] %s: Extracting information' % video_id)
+
+ def _real_initialize(self):
+ # Retrieve disclaimer
+ request = urllib2.Request(self._DISCLAIMER, None, std_headers)
+ try:
+ self.report_disclaimer()
+ disclaimer = urllib2.urlopen(request).read()
+ except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+ self._downloader.trouble(u'ERROR: unable to retrieve disclaimer: %s' % str(err))
+ return
+
+ # Confirm age
+ disclaimer_form = {
+ 'filters': '0',
+ 'submit': "Continue - I'm over 18",
+ }
+ request = urllib2.Request(self._FILTER_POST, urllib.urlencode(disclaimer_form), std_headers)
+ try:
+ self.report_age_confirmation()
+ disclaimer = urllib2.urlopen(request).read()
+ except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+ self._downloader.trouble(u'ERROR: unable to confirm age: %s' % str(err))
+ return
+
+ def _real_extract(self, url):
+ # Extract id and simplified title from URL
+ mobj = re.match(self._VALID_URL, url)
+ if mobj is None:
+ self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
+ return