Debugging

Debugging is the process of testing the logic of your rules to confirm that inputs match expected decisions. Debugging is also used to see how rules, screens, controls and data interact in an interview. The Oracle Policy Modeling Debugger is a tool that can be used to perform this testing function. To get started you should:

• Launch the Policy Modeling Debugger
• Find Your Way Around the Debugger User Interface

Using the Debugger you can:

• Investigate a goal and set the input attributes relevant to that goal as required. This is a top down approach to testing. For more information, see Investigate a Goal and Set the Value for an Input Attribute.
• Set the input attributes and then view which attributes are inferred from those. This is a bottom up approach to testing. For more information, see Set the Value for an Input Attribute and View the Inferred Attributes.
• Create entity and relationship instances to investigate any rules in the Debugger which use those entities and relationships. For more information, see Define Entity and Relationship Instances to Use In a Debug Session.
• See all the relevant paths that contributed to the value of a known attribute. For more information, see View the Decision for a Known Attribute.
• Perform targeted testing of your rules by manually setting attributes and investigating custom goals. For more information, see Test a Portion of a Policy Model.
• Test any data validations you have defined for the interview. For more information, see Test Input Validations and Test Error and Warning Event Rules.
• Check that entity instances are inferred as members of a relationship as intended. For more information, see Investigate an Inferred Relationship.
• Enter and visualize temporal data, and check temporal outcomes. For more information, see Debug Temporal Rules and Data.
• Find the cause of a logic error when your policy model is not operating in the intended way. For more information, see Investigate a Logic Error and Fix a Logic Error.
• Find the cause of a rule loop. For more information, see Investigate a Rule Loop.
• Reuse session data to facilitate retesting after making any necessary fixes to the rules and screens. For more information, see Retain a Debug Session.
• Load a checkpoint saved from a data connector. For more information, see Load Checkpoints Into a Debug Session.
• Draw valid conclusions where attributes or relationships used in a rule have an unknown value. For more information, see Get Answers Even When Relationships Are Only Partially Known.
• Test an interview to ensure that your screens are effective and work as expected. For more information, see Test an Interview.
• Use sample data, that simulates the dynamic data that would be loaded from a connected application, to test rule and interview logic. For more information, see Test Dynamically Loaded Data.
• Test that decision services are invoked as expected. For more information, see Test a Policy Model Containing Decision Service References.