Validating coaches in heritage human services (deprecated)

To ensure that a coach that you use in your heritage human service passes valid data in the flow, use validation to check the coach data.

About this task

In a heritage human service, you validate the data that is in the coach before the flow proceeds to the next step in the service flow. You validate the data by adding a validation element and a stay-on-page event. In a typical implementation, the validation element is a server script, but it could also be a service. If you use a service, the service can be either a heritage service, a service flow, or another human service, and must have a coach validation output variable that matches the coach variable. The server script is the simpler implementation although the service implementation can provide more flexibility.
Note: In the validation path of the heritage human service, you cannot have more than one validation element, which can be either a server script or a service.

The following procedure describes how to validate coach data by using a server script. For an example, see Example: validating coach data in a heritage human service in the web Process Designer.

For information about using a service to validate coach data in the desktop Process Designer, see Example: validating coach data in a heritage human service in the desktop Process Designer (deprecated). For information on how to validate data for shared business objects, see Save services for shared business objects.

Procedure

  1. To validate the coach data before a boundary event occurs, select the connection between the coach and the end event. In the properties for the connection, set Fire Validation to Before.
    The connection shows a green dot at its start, which indicates that validation is enabled. A boundary event Validate is automatically attached to the coach, indicating that you can connect the coach to a validation script.
  2. Add a server script that contains the logic for data validation to the human service diagram.
  3. Select the script of the validation pattern. In the Script properties, add JavaScript code that identifies problematic data.
    Use the tw.system.addCoachValidationError(CoachValidation coachValidation, String errorBOPath, String errorMessage) API to add the error data to the coachValidation system variable.
    For example, you want two fields to have a value. They are bound to var1 and var2. To ensure that they have a value, use code like the following example code:
    if (tw.local.var1 == ""){
	tw.system.addCoachValidationError(tw.system.coachValidation, "tw.local.var1", "Enter a value for field 1");
}
if (tw.local.var2 == ""){
	tw.system.addCoachValidationError(tw.system.coachValidation, "tw.local.var2", "Enter a value for field 2");
}
    The tw.system.coachValidation parameter is the system variable that contains the validation information. The first string contains the full variable path to the data element with the problematic data. The second string is the message for the user. The message identifies what is wrong with the data and tells the user how to fix the problem.
    Important:
    • A coach in a heritage human service can use only one validation script to validate its data. However, more than one coach can use the same validation script.
    • If the data element that is being validated is not bound to a view, there is nowhere to display a validation error if one occurs.
    • If a view that is being validated contains rich text, code the validation script to remove formatting before it validates the text contents.
  4. Connect the validate icon of the coach to the validation script.
    A green dot at the coach end of the connection shows that validation is enabled.
  5. Add a stay-on-page event to the diagram, and then connect the validation script to the stay-on-page event.
    This construct loops the service flow back to the coach if the data in the coach is not valid. At run time, the system passes error information back to the coach and users see an indicator beside the view with the problematic data. If the validation script provides error messages, users see the appropriate message when they hover over an indicator. If the data is valid, the system processes the boundary event to move to the next step.
  6. Click Save or Finish Editing.