DSXchange

My guess...
Enable 'returns multiple rows' in lookup and set the missing option as 'continue'.

Is it you who developed the job? Maybe you can shed more light on what you need to achieve.

I do not understand the reason of restricting view of question as premium content.

I think the o/p implies resetting an aborted job.

But not sure why bothered about when and why.

What happens if you write the data in order of keyfields first followed by data fields? Assuming you have k1, k2, v1, v2, k3, v3, v4......vn - write them to a sequential file as k1,k2,k3,v1,v2....vn. Then read just defining key and one additional column. Then convert the additional column's delimite...

Dnat,

Did you read what Mike suggested? It will work for you.

That more appears as Varchar. Do yourField[1, 4]

If the source is a seq file, describe the full record (all columns) as a single column with different field delimiter.

In the transformation, replace all delimiters with @VM.

Same was as you do in a sequencer if you want to wait only for 0 seconds.

Build a loop within the routine.

But why do you want to build a routine when you have the GUI ?

Try

yourField[1:length(yourField) - 1]

A 'for' loop for the length subtracting 'A' from each char is what you need.

Alternatively have 26 replace commands.

You can further simplify this as

outputString = replace('ab...j','0....9',inputField)

and do the rest 16.

Appears that last semicolon is missing.

Have you defined the columns as string or ustring?

Your requirement drives the answers. Specifying requirement in refined format and in full will lead to better answers. You wanted to wait for 30 mins. Now you want to run 300 such jobs within a day => 300 x 30 mins = 9000 mins sleeping time. Mins in day = 24 * 60 = 1440. So you will need 6.25 days f...

Do you have a decimal or sign? Because PX enforces numeric value to have sign.

Do you have any User-defined SQL?

When you said it works in server jobs, do you imply that you can see more than 6 chars in the field value?