无Script Task ID时，如何为其设置局部变量？

Got it, let's tackle this problem head-on. The core issue here is that when you complete the user task, the subsequent script task hasn't been instantiated yet—Camunda only creates task instances when the flow actually reaches them. That's why you can't get its ID to call taskService.setVariableLocal() directly.

Here are three practical, battle-tested solutions to work around this:

方案1：用BPMN输入参数映射（最简洁） #

This is my go-to for simple scenarios, since it requires minimal code changes and leverages Camunda's built-in functionality.

1. When completing the user task, pass your target data as a process variable using taskService.complete(taskId, variables). Just put the data you need for the script task into the variables map.
2. Open your BPMN file, find the script task, and add an Input Parameter:
   • Set the parameter name to whatever you want your local variable to be (e.g., scriptLocalData)
   • For the value, use an expression that references the process variable you just set (e.g., ${userTaskOutput})

Once the flow moves to the script task, Camunda automatically copies the process variable value into the script task's local scope. This variable will only exist for the script task—no other tasks in the process can access it, exactly what you need. You can use it directly in your script by the parameter name.

方案2：用Script Task的启动监听器处理动态逻辑 #

If you need more dynamic control (like calculating the local variable value based on context), use an Execution Listener on the script task's start event.

1. When completing the user task, store your temporary data in a process variable (e.g., tempScriptInput).
2. In your BPMN definition, add a start event Execution Listener to the script task. You can use a Java class or delegate expression:
   • For a Java class, implement the ExecutionListener interface:

     public class ScriptTaskInitListener implements ExecutionListener {
         @Override
         public void notify(DelegateExecution execution) throws Exception {
             // Grab the temp data from process variables
             Object inputData = execution.getVariable("tempScriptInput");
             // Set it as a local variable for the script task
             execution.setVariableLocal("scriptLocalVar", inputData);
             // Clean up the temp process variable to avoid clutter
             execution.removeVariable("tempScriptInput");
         }
     }
     

3. When the script task starts, this listener runs automatically, converting the temp process variable into a local one and cleaning up after itself.

方案3：直接在脚本内转换变量作用域 #

If you're using a dynamic scripting language (Groovy, JavaScript, etc.) for your script task, you can handle the variable conversion directly in the script itself:

1. Pass your data as a process variable when completing the user task (e.g., userTaskResult).
2. In your script task's code:

   // Convert process variable to script-local variable
   execution.setVariableLocal("localScriptData", execution.getVariable("userTaskResult"));
   // Remove the process variable to keep scope clean
   execution.removeVariable("userTaskResult");
   
   // Now use localScriptData for your script logic
   

This approach skips modifying BPMN listeners or input parameters entirely, keeping all logic contained within the script.

To recap: You can't get the script task ID before the flow reaches it, so we use process variables as a temporary bridge, then convert them to the script task's local scope either via BPMN configuration, listeners, or inline script code. All these methods achieve your goal of having a variable only accessible to the script task.

内容的提问来源于stack exchange，提问作者Mandroid