DSXchange

Lol, you got me. The funny thing is that we were discussing irrational and imaginary numbers earlier yesterday, which is why the square root of -1 was on my mind --- I just mixed up the two concepts.

I think if you call up the after-job job using the command line 'dsjob' instead of the accepted DS/BASIC set of routines you will be successful.

Kavuri, I'm not sure what you are trying to achieve. Do you intend to do a lossless compression of the integer value into a length 12 string? In that case it makes more sense to keep the original 8-bytes and leave them into a fixed 8-byte length string with no conversion (i.e. RAW). Or build up your...

Please explain 'unable'. Does that mean you get an error message? Or that the export file is empty? Can you edit the job? Do you have other problems in the project? The list of questions goes on.

Sure, irrational numbers. Oh, and my bank account balance (I upgraded from Null())

ramjanaki - what you really need is to use the SEARCH button.

The use of indices in UniVerse type database (to which the DS Engine is closely related) is described as part of the IBM UniVerse Documentation . That will give you the technical details on what indices are and how they are manipulated at a very low level. The menu items in DataStage are just calls ...

You cannot get the job back directly from your UNIX level backup, but don't despair - it doesn't take much effort to do so. But you do need a machine running DataStage. 1. Create a new project, giving it any name you want as it is temporary. 2. Restore your original project (perhaps you have already...

If both tables are in the same instance of DB2 you should just use the join SQL. If I have interpreted the question correctly, you might be better off doing a lookup on table DEF for the main input stream ABC.

DataStage doesn't need to be installed from the CD. What are the permissions on the file "/tmp/dsinstall/.dsinstall"?

Kavuri, as the message states, you have a nullable column and your database DDL states that the same column is not nullable. You need to make sure that null values cannot be present in that column. One way is via a modify stage. Another is to use a function like NullToValue(), NullToEmpty(), NullToZ...

It quite definately is empty and not null. Sequential files don't have null columns.

Are you sure that it is getting a null value or just an empty value?

Is this the first column of that datatype? If so, then it is probably because the engine parses left-to-right and stop when it encounters the bad one.

I think your best bet here would be to export the job and write an editor macro (one reason I love EMACS) for the .dsx file to make these changes. I think you can set default per datatype, but not for datatype/length combinations.