SQL & PL/SQL

Hi experts,

I'm using Synchronous Change Data Capture on 11gR2. I'm facing an issue on tracked columns. I need to exclude somes columns from being captured in my change table.

This is my source table:

CREATE TABLE DEPT (Code INTEGER NOT NULL ENABLE,
                    Name VARCHAR2(400 BYTE) NOT NULL,
                    Salary NUMBER(*,2) NOT NULL,
                    CONSTRAINT PK_CODE PRIMARY KEY (Code));

and I only want to see changes on Code, Name but not Salary.

This is my Change Table definition:

BEGIN
   DBMS_CDC_PUBLISH.CREATE_CHANGE_TABLE(
    owner             => 'cdcpub',
   change_table_name => 'dept_ct', 
   CHANGE_SET_NAME   => 'CHG_SET_DEPT',
   source_schema     => 'my_user',
   source_table      => 'DEPT',
   column_type_list  => 'CODE INTEGER, NAME VARCHAR2(400 CHAR)', 
   capture_values    => 'both',
   rs_id             => 'y', 
   row_id            => 'y',
   user_id           => 'y',
   timestamp         => 'y',
   object_id         => 'N',
   source_colmap     => 'y',
   target_colmap     => 'y',
   options_string    => NULL,
   DDL_MARKERS       => 'n');
END;

When I do:

UPDATE DEPT SET Salary=300000 WHERE CODE=77;

I don't want to see a row for this update in my change table. Is there a way to do this with Oracle CDC or changes are only tracked at table level and not at column's...



Thanks in advance!

Edited by: Salah on Sep 20, 2011 10:11 AM