Flat file to Oracle using --- LKM File to Oracle (SQLLDR)

Question

Hello All,
I am planning to replicate simple sqlldr process below in ODI. Two things I cannot do, 1) I cannot modify exisiting tables for creating new sequences 2) We need SEQ column to start with 1 always(tried sequence in odi, didn't work as expected). I am not really good at jython nor java to be able to code "SEQ SEQUENCE (MAX,1) " process in knowledge module(LKM File to Oracle (SQLLDR)) during control file creation. please advice.
thanks in advance.
OPTIONS ( SKIP=0)
LOAD DATA
INFILE 'e:\Hyperion\DRM\Loadfile\wfc\WFC_1GLO_LE.txt'
BADFILE 'E:\Hyperion\ODI\Logs\WFC_1GLO_LE.bad'
DISCARDFILE 'E:\Hyperion\ODI\Logs\WFC_1GLO_LE.dsc'
TRUNCATE INTO TABLE "MDM_STAGE"."WFC_1GLO_WFLE"
FIELDS TERMINATED BY X'9'
TRAILING NULLCOLS
SEQ SEQUENCE (MAX,1),
PARENT,
CHILD,
CHILD_ALIAS,
ATTRIBUTE1,
ATTRIBUTE2,
ATTRIBUTE3)

Answer 1

You must modify LKM/IKM to create new sequence for each table. Typically, you should load the file into a C$ table in LKM and then call the sequence to get new key ids in IKM stage. But you can do this in LKMif you know what you are doing.
Duplicate LKM or IKM (in this case, LKM File to Oracle (SQLLDR)). Add two steps to the LKM before load step. In the first new step, add drop sequence statement like DROP SEQUENCE <%=odiRef.getTable("L", "TARG_NAME", "W")%>SEQ. Mark this step as "Ignore Error" so that load does not fail when sequence would not exist during the first run. In the second new step, create the sequence <%=odiRef.getTable("L", "TARGNAME", "W")%>_SEQ. That way your sequence value will start from 1 always. Be careful during the incremental run though.
Happy debugging.