When you create a Normalizer transformation in the Transformation Developer, you create a pipeline Normalizer transformation by default. When you create a pipeline Normalizer transformation, you define the columns based on the data the transformation receives from a another type of transformation such as a Source Qualifier transformation. The Designer creates the input and output Normalizer transformation ports from the columns you define.

The following figure shows the Normalizer transformation columns for a transformation that receives four sales columns in each relational source row:

Each source row has a StoreName column and four instances of Sales_By_Quarter.

The source rows might contain the following data:

The pipeline Normalizer transformation has an input port for each instance of a multiple-occurring column.

The following figure shows the ports that the Designer creates from the columns in the Normalizer transformation:

The Normalizer transformation returns one row for each instance of the multiple-occurring column:

The Integration Service increments the generated key sequence number each time it processes a source row. The generated key links each quarter sales to the same store. In this example, the generated key for the Dellmark row is 1. The generated key for the Tonys store is 2.

The transformation returns a generated column ID (GCID) for each instance of a multiple-occurring field. The GCID_Sales_by_Quarter value is always 1, 2, 3, or 4 in this example.