Need help on designing Datastage Job

[epaardekooper]

Hi,

We have a requirement as follows.

1. We have a reference table which has the target table names, Column names (concatenated fields) and other business columns.
2. Many input datasets (from target) which has the columns defined for that dataset.
Now we have to look up the input dataset with the reference table (based on the dataset name)and fetch the column names from reference table and then compare the column names with the input dataset and provide the values (kind of snapshot).
We need to build a generic job and it will be running for each and every dataset.

Kindly let me know how can we achieve this.

Can we use RCP and schema file for the datasets? How to compare column names?

Thanks.

[chulett]

Let's start with what your processing job actually needs to do once it has all of this information, something I'm not clear on from your post. Let's say you generate your schema file and build an RCP job - what exactly does it need to do when it runs? That will determine how viable RCP itself is.

[ray.wurlod]

Build the target table name as a constant (perhaps using a Column Generator stage or a Transformer stage) upstream of the Lookup stage.

A generic approach might use a job parameter for the target table name.

[naveen_1985in]

Let's start with what your processing job actually needs to do once it has all of this information, something I'm not clear on from your post. Let's say you generate your schema file and build an RCP ...

Could not read your entire content ..

[ray.wurlod]

Get yourself a premium membership.

At less than 30c (Rs12) per day it's not expensive, and the revenue is devoted to the hosting and bandwidth costs incurred by DSXchange. So you're helping to keep the site alive.