DSXchange

In general, no. The only yes is where the character sets are compatible and, even then, you may need Oconv(string,NLSmapname) type conversions for non-numeric data. What character set are you using on the DataStage input? Do you have any locale-related specificities, such as date formats, currency, ...

Yes. You have a "read as EBCDIC" property in the Sequential File stage, or you can use the Complex Flat File stage for files with complex structures, such as repeating groups or multiple record types.

Version 8.7 has a lot more command line "stuff", including that one.

... then you have to do them one by one.

What exact version are you running?

Parse the comma-delimited field into separate columns using Transformer stage then use Pivot stage.

This is an ideal situation for using QualityStage.

What do you mean by "2*2" and "2*1"?

The answer to your question 1 is "Yes".

The answer to your question 2 is the number of different fastname property values in the configuration file. More than one indicates MPP.

Not really. If you're working in the account there is LISTF. LISTFL will list just those that are local to the account.

I'm not aware of any limit. Certainly I've seen over 2000 columns per row.

You *could* identify them with @INROWNUM but you're not really interested in the row number - all you're interested in is "this row" and "the previous row".

One of the reasons UniVerse was chosen originally for storing objects is that there is no hard and fast requirement for every record to have the same structure. This remains the case in the UniVerse-derived DataStage engine, and is particularly true in the DS_JOBOBJECTS table. For example a "pi...

Options -wait and -jobstatus are mutually antagonistic. Try removing the -wait option from the dsjob command,

Insert then update tries to insert then, if that fails because the row exist, issues an update. Update then insert tries to update then, if that fails because the row does not exist, issues an insert. Choose the one that best meets your requirements - if there will be more inserts than updates prefe...

It can easily be done in the Transformer stage.
Simply calculate the length in the current record and compare with the length in the previous record. If it's smaller, update the min. If it's larger, update the max. Output at end of group.