If a source file contains encoding definition and it has 
syntax errors,
Python prints line of source file in utf-8.
Here's simple example.

# -*- coding: Shift_JIS -*-
print "XXXXX" abcdefg  # "XXXXX" is Japanese ShiftJIS 
string.

With this source file, Python prints a second line as a 
part of error
message in utf-8 encoding. Lines printed here should be 
reconverted into
original encoding. Also, error column that '^' character 
marks is wrong.

Logged In: YES 
user_id=463672

I filed a patch for this bug.
http://sourceforge.net/tracker/?
func=detail&aid=1031213&group_id=5470&atid=305470

Any chance this bug to be fixed in Python 2.4? 

Logged In: YES 
user_id=33168

I'm hoping that someone more familiar with unicode could
take a look at this.  The patch looks ok to me, but I don't
know how to test that it works.  I'm inclined to accept it,
unless I hear otherwise.

Logged In: YES 
user_id=33168

By "it" I meant the patch which is where I intended to make
the previous comment.

#1031213 contains a patch, marking as superseder.