[cx-oracle-users] Unicode support

[Anthony T. <ant...@gm...>]

All,

I just checked in to Subversion support for Unicode. Thanks go to
Amaury Forgeot d'Arc for the patch he provided earlier this year and
to everyone else for their patience. :-) Since I have little
experience with languages other than English and I only have access to
little endian machines I'd appreciate those of you with such
experience or machines to test the new code for me and let me know the
results.

Note that now nchar, nvarchar2 data are now returned as unicode. It is
also possible to bind unicode strings in to cursors. SQL statements
and varchar2 and char data are still returned as encoded strings. Note
that with another new feature in cx_Oracle you can force all strings
to be returned as unicode if you so desire. Below is an example that
will run with the data in the test suite:

import cx_Oracle

def OutputTypeHandler(cursor, name, defaultType, size, precision, scale):
    if defaultType in (cx_Oracle.STRING, cx_Oracle.FIXED_CHAR):
        return cursor.var(unicode, size, cursor.arraysize)

connection = cx_Oracle.Connection("cx_Oracle/dev@eadev")
connection.outputtypehandler = OutputTypeHandler
cursor = connection.cursor()
cursor.execute("select * from teststrings")
for row in cursor:
    print "Row:", row

If you have questions, let me know. If you have feedback, please
provide it. Thanks.

Anthony