I might be asking the same question as it has already been discussed a multiple times, but I did not find a proper answer.
I'm using a Datastage parallel job to load for a siebel identity S_COMMENTS, but one of its field , COMMENT is a free flowing text field that can have end of line character within itself, which is legitimate.
A sample structure of the data from source for this identity is given below.
Code: Select all
USER_ID|COMMENT|FROM_DATE
1-10001|Test Comment line1|2015/02/21
1-10002|Test Comment
line2|2015/02/26
1-10003|Test Comment line3|2015/03/17
Please advise me on how ton handle this scenario in a parallel job.(I'm not conversant with server jobs)
My idea: create a pre job sub routine, and parse the file using a sed/awk command to replace the in-between EOL characters with a fixed character of my choice, and revert back in the downstream stages.
Any better ideas will be highly appreciated.
Thanks.
Warm Regards,
Kumarjit.