DSXchange

So... are you looking for someone to test this in a Server job? That's the only way one could use the pivot stage "which is present in server jobs".

I know it will work if I put 50 x spaces on null field value, but it is not best solution. Because, I may have other field with the length char(3000). What makes you think it's not the "best solution"? As noted, fixed-length means every record is always exactly X number of bytes long and ...

Both stages are in the 8.1 version, and it's not the "server pivot stage" - that's completely different, so perhaps I should say all three stages. It is the original PX pivot stage rather than the exciting new Enterprise version introduced in 8.1. 8) I'll see if I can try the other one tom...

Or do you literally want to recreate the Oracle MONTHS_BETWEEN() function in DataStage? The MONTHS_BETWEEN function calculates the number of months between two dates. When the two dates have the same day component or are both the last day of the month, then the return value is a whole number. Otherw...

So... FILLER_2 redefines the PSD_BLF_UNIT_PRICE_PCT field? How can you include both if one redefines the other? My COBOL days are way behind me but that smells a little fishy to me.

Thank you very much craig for testing .Hope you have given datatype as Bigint in both i/p,o/p of pivot and test data is also Big int ( value greater than int ). Yes indeed. :wink: It might help if you explained your troublesome job design in full - stages, processing, etc so we can perhaps do a mor...

FWIW, built an 8.1 job with the Pivot Enterprise stage using BigInt fields and it works just fine. And I can't imagine any kind of View Data error would be relevant to this particular topic.

Do a Mod() of the total number of months by 12 to get any remaining months in the last partial year. And Floor() should work for the years part, I would think.

Convert both dates to internal format using IConv() and then subtract, you'll have the number of days between them. How you get from days to months is up to your business rules.

You'd have to switch to 'user-defined' DDL for that, I assume, and add the CHAR part yourself. Pretty sure the default when you don't specify either is BYTE.

You're already doing that with your substring notation of [1,255] - that limits it to 255 characters. Your problem is those 255 characters need more than 255 bytes to hold them. Any chance of getting the target table altered? All it takes is a simple MODIFY statement for the column in question and y...

Oracle.

Sure, it's an option when creating the column:

FRED1 VARCHAR2(20 CHAR)
FRED2 VARCHAR2(20 BYTE)

Seems to me you'd have to find out where those ones come from and ask them.

Why are you looking for them? The nature of temporary files are such that they are cleaned up after the job completes because they're... well, temporary and thus no longer needed.