如何在 flink-kafka 中省略空值异常，任何帮助都可以

我正在尝试编写一个代码，当温度高于阈值温度（如代码中定义）时创建警报，但键控流正在产生问题。我是 flink 的新手，也是 scala 的中间人。我需要这段代码的帮助

我几乎尝试了一切

 def main(args: Array[String]): Unit = {
    val TEMPERATURE_THRESHOLD: Double = 50.00

    val see: StreamExecutionEnvironment = StreamExecutionEnvironment.getExecutionEnvironment

    val properties = new Properties()
    properties.setProperty("bootstrap.servers", "localhost:9092")
    properties.setProperty("zookeeper.connect", "localhost:2181")

    val src = see.addSource(new FlinkKafkaConsumer010("broadcast",
      new JSONKeyValueDeserializationSchema(true), properties))


    var data = src.map { v => {
      val loc = v.get("locationID").asInstanceOf[String]
      val temperature = v.get("temp").asDouble()
    //val json = Map("locationID" -> locationID, "temp" -> v.temp)
    //val jsonVal = JSONObject(json).toString()
      (loc, temperature)
    }}
 data = data
      .keyBy(
        v => v._1
      )

{"locationID": "ASK", "temp": 35} {"locationID": "BC", "temp": 45} {"locationID": "CHD", "temp": 55} {"locationID": "RAJ", "temp": 65} {"locationID": "EGY", "temp": 55}

我认为我的元组从某处获取空值，我需要帮助忽略该错误

Error:-

Exception in thread "main" org.apache.flink.runtime.client.JobExecutionException: Job execution failed.
...

    at flinkBroadcast1$.main(flinkBroadcast1.scala:59)
    at flinkBroadcast1.main(flinkBroadcast1.scala)
Caused by: java.lang.Exception: org.apache.flink.streaming.runtime.tasks.ExceptionInChainedOperatorException: Could not forward element to next operator
...
Caused by: org.apache.flink.streaming.runtime.tasks.ExceptionInChainedOperatorException: Could not forward element to next operator
...
Caused by: java.lang.NullPointerException