Python - Unable to connect to 2 different databases like hive and iris in same python program

You can make both connections work in the same Python process, but when using jaydebeapi for two different JDBC drivers, you need to pass the paths to both JDBC jar files together in the connection call. If you pass only one jar for the first connection and then a different one for the second, the second driver class may not be found. [1]

A working pattern is to build a combined jar list like this: [1]

import jaydebeapi as jd

IRIS_JARFILE = "/path/to/intersystems-jdbc.jar"
HIVE_JARFILE = "/path/to/hive-jdbc.jar"

IRIS_DRIVER = "com.intersystems.jdbc.IRISDriver"
HIVE_DRIVER = "org.apache.hive.jdbc.HiveDriver"

JDBC_JARFILES = [IRIS_JARFILE, HIVE_JARFILE]

hive_con = jd.connect(
    HIVE_DRIVER,
    jdbc_conn_url,
    [hive_user, hive_pass],
    JDBC_JARFILES
)

iris_con = jd.connect(
    IRIS_DRIVER,
    iris_conn_url,
    [iris_user, iris_pass],
    JDBC_JARFILES
)

The key point is this:

“note connecting to two data sources in the same program requires that you

reference the paths to BOTH jar files in a list in every call. Otherwise the

second connection attempt will fail. It appears that the paths only get

added once within a process.” [1]

Your current code uses a single jarfile argument per call, which matches the single-database examples for IRIS JDBC connections, but for your two-driver case you should replace that with a list containing both jars. [1][2]

For IRIS specifically, the driver class name shown in the example is: [1][2]

com.intersystems.jdbc.IRISDriver

And the standard JayDeBeApi connection form for IRIS is: [2]

conn = jaydebeapi.connect(driver, url, [user, password], jarfile)

So the fix is not the driver name itself, but changing the last argument from one jar path to the combined list of both driver jars for both connections. [1][2]

Sources: