Tracing Columns

Tracing Columns must exist in all Transaction Pipeline target Data Stores. Data Stores which are used for user input through Worksheets does normally not have Tracing Columns, because the system does not use then in that context.

• SYS_TransGeneratorID The ID of the process that generated the transactions, for example the Spawner or a custom Business Module Data Flow Item.

• SYS_OriginID

The ID of the Data Store that the transactions were originally read or generated from.

• SYS_OriginRowIdentity

The value of PBRowIdentity from the origin Data Store.

• SYS_OriginColumnID

The name of the source column in the origin Data Store.

• SYS_DataSetID

The name of the Data Set being processed. When processing data from an input Data Store (where SYS_DataSetID does not exists, or does not have a value) to a target Data Store, you need to specify how the Data Set Name should be resolved.

The most common ways of doing this are:

- When running a Transaction Pipeline process, for example the Cell Adapter, add an entry to Auxiallary Data of the Input configuration as a **Computed Column** (checkbox on the right side). Call the Computed Column "SYS_DataSetID" and specify the Data Set name in the **Expression**. You can either hard code the value as a single-quoted text, or use the `@Object[<source>].DataSet directive`.

Note

Using the @Object[…].DataSet directive requires that the Data Store is registered in the Data Set configuration of a Workflow, and it must be enclosed in single quotes, ‘@Object[<workflow name>].DataSet’, since it will be used as part of a SQL SELECT clause.

- If you want to load data from one Data Store to another using the Data Store designer, you need to add a binding from the source Data Store to the **SYS_DataSetID** column of the target Data Store and use the `@Object[…].DataSet` directive in the Expression property of the binding.

- Using a custom SQL script, you can use the @Object[…].DataSet directive to resolve the Data Set name of the source Data Store.