[cx-oracle-users] right way to use Unicode under Python 2.7?

[Chris R. <chr...@hu...>]

So, Unicode mode / WITH_UNICODE seems to have been removed a while ago
(https://bitbucket.org/anthony_tuininga/cx_oracle/commits/33157874357bc4209
48b90c7db28865d0138a3e2 ), although it's still referenced in the docs in a
couple places 
(http://cx-oracle.sourceforge.net/html/connection.html#Connection.encoding
<http://cx-oracle.sourceforge.net/html/connection.html?highlight=encoding#C
onnection.encoding> ).

What, then, is currently The Right Way™ to setup cx_Oracle to completely
use Unicode under Python 2.7?

Doesn't seem to be working quite right out-of-the-box (nor are the error
message or traceback particularly helpful):

>>> query = u"SELECT COL_A, COL_B FROM some_db.some_table WHERE COL_A LIKE
>>>:0 ESCAPE '@' ORDER BY COL_A ASC"
>>> param_vals = [u"%☂%"]  # U+2602 UMBRELLA
>>> param_vals
[u'%\u2602%']
>>> cur = connection.cursor()
>>> cur.execute(query, param_vals)
Traceback (most recent call last):
File "<console>", line 1, in <module>
UnicodeEncodeError: 'ascii' codec can't encode character u'\u2602' in
position 1: ordinal not in range(128)



Relevant versions:
Python 2.7.3
cx_Oracle 5.1.2
Oracle Instant Client for Linux x86-64 11.2.0.3.0


Thanks,
Chris Rebert