homepage

Python debugging under console is a PITA, because it has a bad habit to
fail with UnicodeEncodeError in case of unknown encoding in output. It
quickly turns into a headache when inspecting methods like in the
following example running under windows:
>>> import active_directory
>>> users = active_directory.search ("objectCategory='Person'",
"objectClass='User'")
>>> u = users.next()
>>> u = users.next()
>>> u = users.next()
>>> u.dump()
LDAP://CN=LыхъёрэфЁ,CN=Users,DC=dom,DC=com
{
Traceback (most recent call last):
 File "<stdin>", line 1, in <module>
 File "build\bdist.win32\egg\active_directory.py", line 495, in dump
 File "C:\Python25\lib\codecs.py", line 303, in write
 data, consumed = self.encode(object, self.errors)
 File "C:\Python25\lib\encodings\cp1251.py", line 12, in en
code
 return codecs.charmap_encode(input,errors,encoding_table)
UnicodeDecodeError: 'ascii' codec can't decode byte 0x84 in position 8:
ordinal
not in range(128)
Will this be fixed in Py3k to allow range(255) in case of unknown
encoding? Or will there be a new wrapper function - some
rawhexprint(...) that will temporary change sys.output encoding for the
time arguments are executed and process out of range symbols in output
stream according to its embedded rules (i.e. converting to hex
representaton). Another function can be supplied to write raw bytes to
console as-is. Raw write is preferable as it gives the possibility to
redirect output to a file and inspect it later.
These encoding issues in my POV is the stumbling block that makes
scripting in Python 2.x harder for non-english users and hampers overall
Python adoption. Is this going change in Py3k?

I think this had been addressed by PEP 3138 [1].
[1] http://www.python.org/dev/peps/pep-3138/