public interface TaskIO

Task Input / Output

Each task has "inputs", which are copied from the Process state when the task is started, and "outputs", which are atomically applied to the Process state when a task is completed.

Tasks can use Task.inputField to specify the field from the Process state that should be used as inputs, and Task.outputField to specify the field in the Process state that the task should write output.

More complex tasks can take multiple fields from the process state via Task.inputFieldList and write to multiple fields of the process state via Task.outputFieldList. In this case, the task is said to have an "input Record" and/or "output Record", which is a copy of the process state Record with only the fields listed in the inputFieldList copied.

When both inputField and inputFieldList are specified, the inputField is considered the "primary" input field and will be used automatically by various Task subclasses.

An additional option for output is provided in Task.outputExpression to write task output directly into another DataBoundComponent instead of or in addition to the process state. See details below.

inputField and inputFieldList examples

inputData represents the result of the inputField processing and inputRecord represents the result of the inputFieldList.

If the Process.state represented in JSON is:

  {
     orderId:5,
     orderItems: [
        {name:"Pencils", quantity:3, itemId:2344}
     ],
     orderUser: { name:"Henry Winkle", address:"...", ... }
  }
  

Consider these input definitions and resulting inputData and inputRecord:

• inputField: orderId
  • inputData: 5
  • inputRecord: { orderId: 5 }
• inputField: orderUser.name, inputFieldList: [ "orderUser" ]
  • inputData: "Henry Winkle"
  • inputRecord: { name: "Henry Winkle", orderUser: { name: "Henry Winkle", address: ... }
• inputField: orderUser
  • inputData: { name: "Henry Winkle", address: ... }
  • inputRecord: { orderUser: { name: "Henry Winkle", address: ... } }
• inputFieldList: [ "orderUser" ]
  • inputData: null
  • inputRecord: { orderUser: { name: "Henry Winkle", address: ... } }

Notice that inputField is implicitly added to the inputRecord as a field with the same name.

Field paths

Specifying an input or output field as a dataPath allows a hierarchical process state to be flattened into the task input or expanded from the task output.

Output expressions

An Task.outputExpression can be specified to write task output directly into another DataBoundComponent instead of or in addition to the process state.

An output expression is a String prefixed with "$" followed by the DataBoundComponent ID and optionally followed by a dot-separated field name. When no optional field name is specified the task output is written to the target component using setValues() or setData(). With the optional field name, the task output is written to the target with setFieldValue() or setEditValue(). For a ListGrid the row is either the current edit row or the one selected row.

As an example, consider a DynamicForm with ID of "orderHeader". By specifying an outputExpression as "$orderHeader" for a fetch ServiceTask the response record will be assigned directly to the DynamicForm.