Auto Correct Node

The auto correct node sets up the workflow for the auto correction process. The node is to be called when the validation of a transaction fails. It replaces a series of selected fields and returns to the workflow. The workflow must configured to cycle back to the starting point node. If the workflow succeeds, the auto correction node is bypassed and the changed fields are returned to the client. If it fails, the node is called again to try a different set of fields or returns with a failure. A failure alerts the workflow that all attempts to adjust the item failed and all data fields are reset to their original values.

The node uses an initialization table to obtain most of its set up information. The node's set up table is specified in the initTask element, which provides the name of an initialize table type task descriptor.

The auto correction node is a multiple pass node. It has a pass counter associated with it, which is identified by the passCounterFieldName element in the node's XML file. It is a required user data field and the dataType must be byte with an integer value. For workflows with multiple auto correction nodes, the counters are configured separately.

The node can change the value of the NPR MICR fields, but is not limited to this set. It can be used to change other fields. When NPR MICR fields are used, the auto correction node sets the ibmNprAutoCorrected flag and returns it to the client. If the client is the Gateway, it is routed to the FTM database. The value is yes if auto correction was successful and no if the attempt failed. The value is blank if auto correction was not attempted. The corresponding valid flag for the MICR field is adjusted accordingly.

On the first call to the auto correction node for an item, all modified data fields in the workflow are backed up before the specified fields are replaced. On subsequent calls, the node replaces the backed up fields before trying a new set or returning a failure.