Modify Stage specification Parameterization

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
nani0907
Participant
Posts: 155
Joined: Wed Apr 18, 2007 10:30 am

Modify Stage specification Parameterization

Post by nani0907 »

Hi All,

I had parmeterized the modify stage specification to make my multi-instance job generic and reusable.In one of the execution, the job does not require the paramater. In that case i didn't pass anything to job parameter,It aborts.Could any body help what to pass in specification parameter if there in no transformation required.

Thanks in advance
thanks n regards
nani
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

Could you give an example of what the parameter looks like? Perhaps you might need to add a dummy modify stage action such as dropping a non-existant column.
nani0907
Participant
Posts: 155
Joined: Wed Apr 18, 2007 10:30 am

Post by nani0907 »

Hi sir,

It's a job parameter where iam using for dropping columms and changing datatypes.For one of the run, i have my source metadata &target metadata with same layout which doesn't require that job paramater in modify stage.How to proceed in that scenario...
thanks n regards
nani
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

Could you give an example of what the parameter looks like?
nani0907
Participant
Posts: 155
Joined: Wed Apr 18, 2007 10:30 am

Post by nani0907 »

Here is the value i passed for my first multi-instance run and complted sucessfully--- drop A_DUMMY;drop B_DUMMY; IC_ID : string [16] = IC_ID

In second multi-instance run, i not passing any thing to parameter as there are no transformation required.Pls suggest..
thanks n regards
nani
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

Try passing "drop nothing" as the parameter.
surfsup
Participant
Posts: 18
Joined: Thu Apr 23, 2009 8:43 am

Post by surfsup »

It's unlikely the source data will have a column called nothing, but "nowarn" also works.
BI-RMA
Premium Member
Premium Member
Posts: 463
Joined: Sun Nov 01, 2009 3:55 pm
Location: Hamburg

Post by BI-RMA »

ArndW wrote:Try passing "drop nothing" as the parameter.
Usually if You drop a nonexisting column I would expect Datastage to throw a warning telling You that column "nothing" is not present in the input schema. To prevent this You will either have to demote this warning to informational (with the risk that you miss messages related to other columns), or add a Dummy-column to the stream that you can safely drop within the modify stage.

A modify stage with an empty specification is invalid.
"It is not the lucky ones are grateful.
There are the grateful those are happy." Francis Bacon
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

I suggested dropping a nonexistant column in the OP's modify stage as a route to solving the problem. As you noted, the warning message would need to be deprecated. But with that approach, the OP could have added a field that would get dropped in the modify, or done a "{col}={col}" where "col" is known column name that is always in the data. As you noted, the core of the problem was than an empty command is not allowed.
Post Reply