Hi all,
There's a requirement were, a large file has to be inserted into the database daily. This job is been scheduled. The output component is tPostgresqlBulkExec. But using this component we can either make Bulk insert or Bulk update, but not Both. Since there's large number records, I cannot use PostgresqlOutput, because using this component it takes it takes long time to execute. I need to make insert/update (to avoid duplicate rows).
Is there any way to do this?
Hi,
You should use a tMap and check if the key are already present in the Target Database.
You will have two outputs from the tMap:
Output 1, will have rows to insert, linked to a bulk insert
Output 2, will have rows to update, linked to an update component.
regards,
Benjamin
Hi,
You should use a tMap and check if the key are already present in the Target Database.
You will have two outputs from the tMap:
Output 1, will have rows to insert, linked to a bulk insert
Output 2, will have rows to update, linked to an update component.
regards,
Benjamin
Thanks Benjamin, can you guide me with the function that has to be used in tmap regards, Swathi
No a existing function can be used, you need to query the existing key from target database and do a inner job on tMap, get the matched rows for updating, the unmatched rows for inserting. eg:
tFileInputDelimited--main--tMap-----main--tPostgresqlBulkExec(for inserting)
| -----main--tPostgresqlBulkExec(for updating)
lookup
|
tPostgresInput(select key_column from target_database)
How to do a inner job on tMap and get matched/unmatched rows? see
https://community.talend.com/t5/Design-and-Development/tJMSInput-configurations-for-IBM-MQ-messaging... Best regards
Shong
