[collegehumor] Encode the xml before calling xml.etree.ElementTree.fromstring (fixes #1822)

Uses a new helper method in InfoExtractor: _download_xml

1 files changed, 6 insertions, 0 deletions

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 3cebeaf29..482a231ec 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -4,6 +4,7 @@ import re
 import socket
 import sys
 import netrc
+import xml.etree.ElementTree
 
 from ..utils import (
     compat_http_client,
@@ -208,6 +209,11 @@ class InfoExtractor(object):
         """ Returns the data of the page as a string """
         return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
 
+    def _download_xml(self, url_or_request, video_id, note=u'Downloading XML', errnote=u'Unable to downloand XML'):
+        """Return the xml as an xml.etree.ElementTree.Element"""
+        xml_string = self._download_webpage(url_or_request, video_id, note, errnote)
+        return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))
+
     def to_screen(self, msg):
         """Print msg to screen, prefixing it with '[ie_name]'"""
         self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))