Cython ODBC library
Very incomplete and very incompatible with ODBC and/or Python DBAPI 2.0
Might never be compatible with anything
Use at own risk and everything
Currently it only supports executing a simple SELECT statement without
any parameters.
Primary goal is to be very fast with fetching lots of rows, especially if you
set cursor.arraysize to a reasonable amount such as 100 before you run
cursor.execute()
SELECT 20,000 rows with 10 columns of different column types
(varchar, int, bigint, double) from a MySQL server running on localhost,
with cursor.arraysize set to 100:
pyodbc
average time (of 10 runs): 3.5033s
ceODBC
average time (of 10 runs): 0.0647s
cyodbc
average time (of 10 runs): 0.0649s