Patch #1309009, Fix segfault in pyexpat when the XML document is

in latin_1, but Python incorrectly assumes it is in UTF-8 format

Will backport.

1 files changed, 9 insertions, 0 deletions

diff --git a/Lib/test/test_minidom.py b/Lib/test/test_minidom.py
index 3154fbf..8b4c715 100644
--- a/Lib/test/test_minidom.py
+++ b/Lib/test/test_minidom.py
@@ -889,6 +889,15 @@ def testEncodings():
             and doc.toxml('utf-8') == '<?xml version="1.0" encoding="utf-8"?><foo>\xe2\x82\xac</foo>'
             and doc.toxml('iso-8859-15') == '<?xml version="1.0" encoding="iso-8859-15"?><foo>\xa4</foo>',
             "testEncodings - encoding EURO SIGN")
+
+    # Verify that character decoding errors throw exceptions instead of crashing
+    try:
+        doc = parseString('<fran\xe7ais>Comment \xe7a va ? Tr\xe8s bien ?</fran\xe7ais>')
+    except UnicodeDecodeError:
+        pass
+    else:
+        print 'parsing with bad encoding should raise a UnicodeDecodeError'
+
     doc.unlink()
 
 class UserDataHandler: