DSXchange

Without knowing more I would suggest sticking with an aggregator to get the minimum of columns A and B, then going to a transform and output those two column plus the MIN of A and B minimums. The transform only gets executed one time, since there is only one row. This is better than your current sol...

If all you need is one MIN value across all value in ColA and ColB then all you need is one stage variable to hold that. You would only need to do one pass through the data; any sort or aggregation is going to have more processing to do than that. The question then is what you plan on doing with tha...

Glad to hear you were able to solve your problem, even though I can't see it works differently. I trust you are certain that it is now working correctly.

The more efficient way to do this would be to use just one stage, the Aggregator, it will perform not only the MIN but also the MAX at the same time for you.

Change your job to write out the value to your output file and see what the value really is, this might help you more than knowing what it isn't. The value might be a non-displayable one, not a "" or a SQL.NULL or CHAR(000).

What happened when you tried using ""? Also, using isnull(stagevar) does work [I just tried it] so you must have some other sort of error elsewhere in you logic.

This is because functions need to be pre-declared in programs using the DEFFUN declaration.

In your case the solution is simple, create a routine activity stage and call up your function, then in the Nested Condition use the value Your_Routine_Activity.$ReturnValue" for your comparision.

If you do not have active-to-active stage interprocess buffering enabled then all the transforms will be executed by one process, and row 2 won't be read until row 1 processed; so there is no need for passing global values between stages. I don't understand why you have this restriction in your job ...

You can access a link's value using @INROWNUM in the same transform. If you wish to use that value in another transform downstream, then add this value to a new column in the output derivation. I'm not sure I understood your question or problem, but this might answer it. If not, perhaps you could ex...

The null value is quite different from the "" (empty string value). In your case the FIELD() function will return the empty string, so your derivation should read: If (Code = "A" or Code = "B" or Code ="C" or Code ="") Then "Valid" Else "...

babbu9,

I think that Ray and I have understood that; but when you use a transform stage the compiler will do several additional actions. Have you checked your access to this directory (and to those above it in the path)? It might be that the default umask settings for this user aren't correct.

sobral,

what happens if you use ldd or odump on this (/home/dsadm/Ascential/DataStage/DSEngine/lib/libvmdsapi.so ) library file? Is it detected and displayed as a valid library?

Do you still have the problem if you make your sequence call only the one job and do nothing else?

Tamas, a quick look at the DSD.GetStatus routine shows that it is accessing the file variable (the job's status file). The error message means that this file has not been "opened" or that somehow else the file pointer has been invalidated. This file pointer is set as a function of the DSAttachJob() ...

Ganive - excellent solution; somehow I got this thread mixed up with the other one currently running with a similar subject content and recalled that the $APT_STRING_PADCHAR didn't work so I didn't even consider that approach. You've got the best solution for your issue now, though.