Thread: Executing stored procs

Hi,

I’ve been currently trying to execute stored procs with psycopg2 on an AWS redshift cluster but no matter what I do I can’t figure out the syntax for how to get it to successfully execute.

CREATE OR REPLACE PROCEDURE stg_customervip.sp_STG_Customer ( 

                                                                    p_Audit_RunID IN bigint ,

                                                                    p_Audit_RunDTS IN timestamp

                                                                    )

AS $$

BEGIN

TRUNCATE table stg_customervip.Customer;

INSERT INTO stg_customervip.Customer

       (

        ,Audit_RunID    

        ,Audit_RunDTS

        )

SELECT 

        ,p_Audit_RunID  AS Audit_RunID

        ,p_Audit_RunDTS AS Audit_RunDTS

FROM stg_customervip_temp.customer;

END;

$$ LANGUAGE plpgsql; 

try:

    redshift_conn = psycopg2.connect(dbname=args.redshift_database_name, host=args.redshift_address,

                                        port=args.redshift_port, user=redshift_creds['username'], password=redshift_creds['password'])

except Exception as e:

    raise(e)

rand = random.randrange(

    15, 45)

timestamp = datetime.utcnow().strftime("%Y-%m-%d %H:%M:%S")

for proc in stored_procs_obj:

    qry_str = sql.SQL("CALL {proc}({int}::bigint, {timestamp}::timestamp)").format(

        proc=sql.Identifier(proc['schemaname'], proc['procedurename']), int=sql.Literal(rand), timestamp=sql.Literal(timestamp))

    logger.info(qry_str.as_string(redshift_conn))

    logger.info(redshift_conn.closed)

    redshift_conn.reset()

    try:

        with redshift_conn.cursor() as cursor:

            cursor.execute(qry_str)

            redshift_conn.commit()

            logger.info("Query successfully committed")

    except Exception as e:

        redshift_conn.close()

        send_sfn_failure(args.task_token, "Exception", str(e))

        raise

Essentially results in a SQL query of: CALL "stg_customervip"."sp_stg_customer"(37::bigint, '2020-01-15 05:52:31'::timestamp)

If I take that query and run it directly in redshift it runs successfully but via psycopg2 it returns:

psycopg2.errors.FeatureNotSupported: TRUNCATE cannot be invoked from a procedure that is executing in an atomic context.

HINT: Try calling the procedure as a top-level call i.e. not from within an explicit transaction block. Or, if this procedure (or one of its ancestors in the call chain) was created with SET config options, recreate the procedure without them.

CONTEXT: SQL statement "TRUNCATE table stg_customervip.Customer"

While that error sounds legitimate I think it might be a redherring because it does execute successfully if I do it manually against redshift.

I’ve also attempted to try: SELECT "stg_customervip"."sp_stg_customer"(37::bigint, '2020-01-15 05:52:31'::timestamp) but that returns: [Amazon](500310) Invalid operation: stg_customervip.sp_stg_customer(bigint, timestamp without time zone) is a procedure;

I did notice there was a callproc() function available to call a function or a procedure but that doesn’t make it any further

        param1 = str(rand) + "::bigint"

        param2 = timestamp + "::timestamp"

        try:

            with redshift_conn.cursor() as cursor:

                cursor.callproc(proc['fullprocedurename'], (param1, param2))

                redshift_conn.commit()

                logger.info("Query successfully committed")

        except Exception as e:

            redshift_conn.close()

            send_sfn_failure(args.task_token, "Exception", str(e))

            raise

returns:

psycopg2.errors.WrongObjectType: stg_connect.sp_stg_store("unknown", "unknown") is a procedure

HINT: To call a procedure, use CALL.

Has anyone encountered this or has any suggerstions?

Thanks

Ah awesome! Thanks Daniele that appears to have worked!

Thanks

From: Daniele Varrazzo <daniele.varrazzo@gmail.com>
Sent: Wednesday, 15 January 2020 10:06 PM
To: Anthony Waye <anthony.waye@arq.group>
Cc: psycopg@postgresql.org
Subject: Re: Executing stored procs