DSXchange

Maybe you can concat them into a single column with comma sep. That way one job will do.

Start your job monitor. Search for the error message.

If it is in seq file (source), you can use awk in the first stage itself to re-organize

If it is in seq file (source), you can use awk in the first stage itself to re-organize

This can be resolved by having a calendar table

Just to add to the logic.....you need to check whether the 'dd' parts are different.

i.e. 2001-01-02 to 2001-02-01 is not one month (if my understanding of your requirement is correct)

So you can add
+ (if second_dd > first_dd then -1 else 0)

The warning is not from oracle but from DataStage.

Check the column formats within the job and in the stage mentioned by the warning.

Even though I am not sure why you need that, a join must meet your requirement.

Or simply run file2 followed by file1....if loading both does not cause any harm !!!!

Please 'search' to find loads of answers.

Check this out

http://www.openaccesssoftware.com/produ ... access.asp

Varchar and fixed width?

There must be some place where you fix the length in output.

I believe if the input is <0 then do not disturbt else you can replace leading blank with '+'.

Write a routine and call from sequence

Alternatively use OS features

Use ODBC or save as csv