Update the columns at run time when you expect small changes to the source. For example, you need to process a source from another organization, but the organization cannot guarantee the order of the columns in the source file. When you enable the option to update the data object columns at run time, the Data Integration Service changes the ports of the Read transformation based on the structure of the source data. The Read transformation passes the data to the downstream transformations in the dynamic mapping for processing.

When you update the data object columns at run time, the Data Integration Service updates a run-time instance of the Read transformation. It does not update metadata in the Model repository. To see the changes in the Developer tool, view the mapping with resolved parameters. To update the physical data object definition in the Model repository, use the synchronize option in the Developer tool. The Developer tool re-imports the physical data object metadata and changes the metadata.

The following image shows where to enable the option on the

Data Object

tab:

The Data Integration Service determines the structure of a relational source by the schema. It examines the schema of the resource that appears on the

Run-time

tab. The Data Integration Service then updates the columns in the transformation data object based on the schema.

The Data Integration Service determines the structure of a flat file source based on the way you configure the flat file physical data object. You configure the data object to generate column names at run time.

Configure this functionality on the

Data Object

tab of the Read transformation for a flat file or relational source.

For more information about setting the flat file physical data object properties, see the

Informatica Developer Tool Guide

.