I have a parallel job which gets its data from a Exec command in Sequence job.
The output of the Exec cmd (lets say A) is appended to the data in the
text file. Delimiter is | . The text file can have 2 records. So A is
appended at the end of both the records.
The Sequence file in the parallel job will be reading 5 files based on pattern and A
will be different in each file.
I need to use a transformer compulsory to see that the A is not
repeated more than once. (Sample data of A = Nov 16 08:24)
The steps I did:
I defined A column as varchar in Sequence file.
In transformer: right(A,1)= d (stage var)
if d = c then 'Y else 'N' = b (stage variable)
d = c (stage variable)
Constraints: b='N'
I not able to get the right output. It is taking all records as
unique. I can see the data in A column. When I say A = c, it is
assigning the value to c but it is not able to compare the data. I am
able to see the data present in c in the output.
If I hardcoded the comparison value then I am able to see the right output.
Eg. if in transformer: i give : if d='4' then 'Y' else 'N'
(if last char in the string is 4)
I am not sure why it is not comparing d & c.
d&c are defined as varchar. If I give AsInteger(right(A,1)) , it shows
transformer error while compiling. Since A is the output from Unix cmd
will there be any different datatype for that?
Comparing stage variables in transformer
Moderators: chulett, rschirm, roy
-
- Participant
- Posts: 38
- Joined: Wed Jul 18, 2007 12:23 am
- Location: Chennai
If you have doubt on-
Or
Please let me know if it is not helpfull.
Then you can apply the below on the execute command outputSince A is the output from Unix cmd
will there be any different datatype for that?
Code: Select all
Trim(Convert(@FM,"",ExecuteCommand1.CommandOutput))
Code: Select all
tr -d '\n'
Please let me know if it is not helpfull.